Fire fighting access occupation identification system and method under cooperation of laser and radar

Abstract

The invention provides a fire fighting access occupation identification system and method under cooperation of laser and radar, and relates to the technical field of laser radars, and the system comprises a point cloud generation module which is used for transmitting a laser beam to a target area through a solid-state laser radar, receiving a reflection signal, extracting a time difference and a phase difference, converting the time difference and the phase difference into polar coordinate data, and sending the polar coordinate data to a server; generating a point cloud data set; the three-dimensional space model building module is used for carrying out point cloud filtering, clustering and three-dimensional reconstruction based on the point cloud data set and building a three-dimensional space model of the target area; and the occupation judgment module is used for calculating a space occupancy rate according to the three-dimensional space model, and executing self-adaptive occupation judgment and pushing alarm information when detecting that the space occupancy rate exceeds a preset dynamic threshold value. According to the method and the device, the technical problem of low pressure occupation detection accuracy in the prior art can be solved, and the technical effects of improving the detection stability, reducing the false alarm rate and enhancing the system adaptability and linkage response capability are achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of laser radar, in particular to a fire passage occupation recognition system and method under laser and radar cooperation. BACKGROUND

[0002] Under the background that urban safety management and intelligent building operation and maintenance are increasingly valued, the smoothness of fire passages becomes one of the key factors to guarantee personnel evacuation efficiency and emergency response capability. Therefore, relevant systems have begun to introduce monitoring solutions based on laser radars, cameras and other devices to identify and alarm the occupation situation in the passage area.

[0003] At present, existing image recognition means are limited by changes in lighting conditions, are easily disturbed by occlusion or shadows, and lack accurate judgment ability of object volume and material, making it difficult to distinguish between temporary personnel passing and continuous occupation behavior. While some detection systems based on single-line or multi-line laser radars can perceive spatial information to some extent, they often ignore environmental noise, point cloud sparsity, sensor dead angles and other factors in actual application, resulting in unstable occupation judgment results and lack of adaptive ability.

[0004] In summary, the existing technology has the technical problem that due to the single sensing mode and lack of environmental adaptation mechanism, the occupation detection accuracy is not high, false positives and false negatives occur frequently, further affecting the real-time and reliability of passage safety management. SUMMARY

[0005] The purpose of the present application is to provide a fire passage occupation recognition system and method under laser and radar cooperation, to solve the technical problem in the prior art that due to the single sensing mode and lack of environmental adaptation mechanism, the occupation detection accuracy is not high, false positives and false negatives occur frequently, further affecting the real-time and reliability of passage safety management.

[0006] In view of the above problems, the present application provides a fire passage occupation recognition system and method under laser and radar cooperation.

[0007] In the first aspect, the present application provides a fire passage occupation recognition system under laser and radar cooperation, comprising: a point cloud generation module for emitting a laser beam to a target area by a solid-state laser radar, receiving a reflected signal, extracting a time difference and a phase difference, converting polar coordinate data, and generating a point cloud data set; a three-dimensional space model construction module for point cloud filtering, clustering and three-dimensional reconstruction based on the point cloud data set to construct a three-dimensional space model of the target area; an occupation judgment module for calculating a space occupation rate according to the three-dimensional space model, and performing adaptive occupation judgment and pushing alarm information when detecting that the space occupation rate exceeds a preset dynamic threshold.

[0008] Preferably, the fire passage occupation identification system under laser and radar synergy is further configured to: a radar synergy unit configured to assign a unique phase offset to each of the solid-state laser radars, configure the emission time of the laser beams according to the unique phase offset, and reassign the unique phase offset if a phase conflict is detected.

[0009] Preferably, the fire passage occupation identification system under laser and radar synergy is further configured to: a space occupation data calculation unit configured to calculate space occupation data based on the three-dimensional space model; and a space occupation rate calculation unit configured to define a reference space of a target area and calculate a space occupation rate in combination with the space occupation data.

[0010] Preferably, the fire passage occupation identification system under laser and radar synergy is further configured to: a contour feature extraction channel configured to distinguish static environment and dynamic occupation objects from the three-dimensional space model using a semantic segmentation algorithm and extract contour features of the occupation objects according to the semantic segmentation result; and a first space occupation rate calculation channel configured to calculate a space occupation rate based on the contour features.

[0011] Preferably, the fire passage occupation identification system under laser and radar synergy is further configured to: a second space occupation rate calculation channel configured to divide an effective passage area from the reference space of the target area and calculate a space occupation rate of the space occupation data based on the effective passage area.

[0012] Preferably, the fire passage occupation identification system under laser and radar synergy is further configured to: a comparison unit configured to compare the space occupation rate with a preset dynamic threshold of space occupation, the preset dynamic threshold being adaptively adjusted according to environmental noise and point cloud density; and an update triggering unit configured to trigger an update of the preset dynamic threshold and retain the space occupation rate when the space occupation rate is greater than the preset dynamic threshold and the occupation duration is greater than a preset anti-transient interference time.

[0013] Preferably, the fire passage occupation identification system under laser and radar synergy is further configured to: an occupation event acquisition channel configured to acquire an occupation event when the space occupation rate is greater than the preset dynamic threshold and the occupation duration is greater than the preset anti-transient interference time after the update; and a reality confirmation channel configured to confirm the reality of the occupation event through a multi-device synergy verification mechanism and trigger an alarm according to a confirmation result.

[0014] Preferably, the fire passage occupation identification system under laser and radar synergy is further configured to: a multi-modal alarm layer configured to perform multi-modal alarm based on the occupation event, the multi-modal alarm including local alarm and cloud alarm.

[0015] Preferably, the fire passage occupation identification system under laser and radar cooperation is further used for a multi-modal alarm sub-layer, wherein the local alarm includes an audible and visual warning, and the cloud alarm includes a three-dimensional position of the occupying object and a material classification result based on reflection intensity.

[0016] In a second aspect, the application further provides a fire passage occupation identification method under laser and radar cooperation, comprising: emitting a laser beam to a target area by a solid-state laser radar, receiving a reflection signal, extracting a time difference and a phase difference, converting into polar coordinate data, and generating a point cloud data set; performing point cloud filtering, clustering and three-dimensional reconstruction based on the point cloud data set to construct a three-dimensional space model of the target area; calculating a space occupation rate according to the three-dimensional space model, and when detecting that the space occupation rate exceeds a preset dynamic threshold, performing adaptive occupation judgment and pushing alarm information.

[0017] The technical solutions provided in the application have at least the following technical effects or advantages: by achieving the technical target of high-precision occupation identification and intelligent alarm based on multi-source data fusion and dynamic threshold judgment, the technical effects of improving detection stability, reducing false alarm rate, enhancing system adaptability and linkage response capability are achieved.

[0018] The above description is only a summary of the technical solutions of the application. In order to enable one skilled in the art to better understand the technical means of the application, and to implement the application according to the content of the description, and in order to make the above and other purposes, characteristics and advantages of the application more obvious and easy to understand, the following specific embodiments of the application are described. It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the application, nor is it intended to limit the scope of the application. Other features of the application will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are only exemplary, and other drawings can be obtained by those skilled in the art without creating any inventive labor on the basis of the provided drawings.

[0020] Figure 1 FIG. 1 is a structural schematic diagram of the fire passage occupation identification system under laser and radar cooperation of the application.

[0021] Figure 2 FIG. 2 is a flow schematic diagram of the fire passage occupation identification method under laser and radar cooperation of the application.

[0022] Legend of the drawings: point cloud generation module 1, three-dimensional space model construction module 2, occupation judgment module 3. DETAILED DESCRIPTION

[0023] The application provides a fire passage occupation recognition system and method under laser and radar cooperation, which solves the technical problem in the prior art that due to single sensing mode and lack of environmental adaptation mechanism, the occupation detection accuracy is not high, false positives and false negatives occur frequently, and the real-time and reliability of passage safety management are further affected. The technical goal of high-precision occupation recognition and intelligent alarm based on multi-source data fusion and dynamic threshold judgment is achieved, and the technical effects of improving detection stability, reducing false positive rate, and enhancing system adaptability and linkage response capability are achieved.

[0024] Hereinafter, the technical solutions in the application will be described clearly and completely with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the application, not all embodiments of the application, and it should be understood that the application is not limited to the example embodiments described here. Based on the embodiments of the application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of protection of the application. In addition, it should be noted that, for convenience of description, only parts related to the application are shown in the drawings, not all.

[0025] Embodiment one, please refer to the accompanying Figure 1 The application provides a fire passage occupation recognition system under laser and radar cooperation, specifically comprising:

[0026] A point cloud generation module 1 is configured to emit laser beams to a target area through a solid-state laser radar, receive reflected signals, extract time difference and phase difference, convert polar coordinate data, and generate a point cloud data set.

[0027] Specifically, the solid-state laser radar is a detection device using a fixed laser transmitter. The solid-state laser radar emits laser beams to the fire passage area to be monitored, such as near-infrared light with a wavelength of 905 nanometers. When the laser beam encounters an obstacle, it will reflect. The receiver inside the radar will capture the reflected signal. After receiving the reflected signal, time difference and phase difference measurement are performed. The time difference refers to the time interval between laser emission and reception. The distance information can be obtained by converting the laser speed, such as a 10 nanosecond time difference corresponding to a 1.5 meter distance. The phase difference is obtained by comparing the waveform difference between the emitted wave and the reflected wave, which can detect displacement.

[0028] The collected time difference and phase difference are converted into polar coordinate data. The polar coordinate system is a coordinate system that uses angle and distance as parameters. For example, a point can be represented by a distance of 5 meters and an angle of 30 degrees. In fire passage monitoring, the polar coordinate can intuitively reflect the position relationship of the obstacle relative to the radar. The polar coordinate data generates a point cloud data set. The point cloud is a set of three-dimensional coordinate points, and each point represents a specific position in space. For example, a standard fire passage monitoring system can generate about 30,000 data points per second, which collectively outline the three-dimensional scene in the passage.

[0029] The three-dimensional space model construction module 2 is used for point cloud filtering, clustering and three-dimensional reconstruction based on the point cloud data set to construct a three-dimensional space model of the target area.

[0030] Specifically, in fire passage monitoring, the data points in the point cloud data set may contain real obstacle information, and may also be mixed with invalid points generated by dust, raindrops or sensor noise. In order to obtain accurate analysis results, point cloud filtering processing is performed. The filtering algorithm will eliminate abnormal points that obviously deviate from the normal range based on parameters such as the spatial distribution characteristics and reflection intensity of the points. For example, a single point fifty meters away from the radar is likely to be a false detection, while a point group densely distributed within a two-meter range is likely to be a real obstacle. After filtering, clustering analysis is performed on the remaining valid point cloud. Clustering is a process of dividing spatially close points into the same group, such as using the Euclidean clustering algorithm to search for adjacent points within a radius of thirty centimeters centered on each point, and connecting the points to the same object. Through point cloud filtering and clustering, the originally disorganized thirty thousand data points may be aggregated into a dozen meaningful object clusters, each cluster corresponding to an independent object in the passage, such as a garbage can or a trolley. According to the clustering results, three-dimensional reconstruction is performed to convert the discrete point cloud data into a continuous surface model, and according to the spatial distribution characteristics of the points, the geometry and surface texture of the object are reconstructed. In the fire passage scene, the reconstructed three-dimensional space model can clearly show the contour features of the obstacle. The constructed three-dimensional space model is a digital virtual scene that accurately reproduces the actual situation of the target area, not only containing geometric information of the object, but also being able to add features such as material properties.

[0031] The occupancy judgment module 3 is used for calculating the space occupancy rate according to the three-dimensional space model, and when it is detected that the space occupancy rate exceeds a preset dynamic threshold, adaptive occupancy judgment is performed and alarm information is pushed.

[0032] Specifically, the three-dimensional space model is a digitized three-dimensional scene established after laser radar scanning, which accurately reproduces the spatial distribution and geometric characteristics of all objects in the monitoring area. It not only contains shape information of the object surface, but also reflects the reflection characteristics of different materials to laser, such as metal surface will produce stronger echo signal than plastic surface. In the fire access scene, the three-dimensional space model can clearly show whether there are illegally parked vehicles or piled up debris in the fire access. The space occupancy rate is a quantitative indicator representing the ratio of the space occupied by obstacles to the total available space of the fire access. By first defining the effective passing area of the fire access, such as a range of six meters long, two meters wide, and two and a half meters high, then counting the volume proportion of the range occupied by the object. If an object with a length of two and a half meters, a width of zero point seven meters, and a height of zero point nine meters is detected, it will be found that it occupies about 5% of the space volume of the channel after calculation. The preset dynamic threshold is a warning value set according to the actual application requirements, which will be automatically adjusted according to the environmental conditions. For example, when there are frequent personnel activities during the day, the alarm threshold may be set to 40%, while at night when the flow is sparse, it is adjusted to 30%. The dynamic adjustment mechanism effectively avoids false positives or false negatives caused by changes in the scene. The basis for setting the preset dynamic threshold includes historical data statistics, real-time environmental monitoring, and management requirements, etc. When the current space occupancy rate is detected to exceed the dynamic threshold, the adaptive occupation judgment is triggered, and various factors are comprehensively considered for judgment, such as the duration of occupation, the material type of the object, and the moving track, etc. A carton appearing in the channel for a short time may not trigger an alarm, but if a motorcycle stays for more than five seconds, it will be judged as an effective occupation event. The final alarm information is the integrated multi-dimensional data, which not only contains the basic occupation alarm, but also attaches detailed on-site situation description. The alarm information may include the three-dimensional size of the occupied object, the location coordinates, the duration, and other key data. At the same time, sound and light alarm devices may be started, such as playing voice prompts to please not occupy the fire access, and flashing warning lights, to ensure that the alarm can be handled in time and effectively.

[0033] Further, the application also includes: a radar coordination unit for assigning a unique phase offset to each solid-state laser radar in the solid-state laser radar, configuring the emission time of the laser beam according to the unique phase offset, and reassigning the unique phase offset if a phase conflict is detected.

[0034] In particular, solid-state laser radar is a kind of detection equipment using fixed laser transmitters. In the fire access monitoring system, multiple radar devices are deployed to achieve all-around coverage. In order to solve the risk of signal mutual interference, each radar is assigned a phase offset, which determines the specific time point of the laser pulse emitted by each radar. The unique phase offset ensures that the laser pulses of different radars are emitted at different times. For example, the first radar may emit at zero seconds, the second radar delays one hundred nanoseconds, and the third radar delays two hundred nanoseconds. Through time control, multiple radars can work simultaneously without interfering with each other. Each radar has a clock inside, which configures the laser beam emission time according to the assigned unique phase offset, can control the emission time, and ensure that each radar emits laser beams according to the plan. Phase conflict refers to the situation where the laser pulses of two or more radars overlap in time. This situation may be caused by clock drift, environmental temperature changes, or equipment failure, etc. When a phase conflict is detected, such as when the emission time difference between two radars decreases from one hundred nanoseconds to five nanoseconds, a reassignment mechanism is triggered, and a new optimal offset combination is calculated according to the current network conditions and device performance. Reassigning the unique phase offset is a dynamic optimization process. Through the working state of each radar, the degree of environmental interference, and monitoring requirements, etc., a new phase configuration is recalculated and issued. For example, in a high-temperature environment, the emission interval between adjacent radars may be increased to cope with the possible slight drift of the clock circuit, ensuring that the monitoring work does not be interrupted. Table 1 is a partial record table of solid-state laser radar phase offset assignment and conflict detection.

[0035] Table 1: Partial record table of solid-state laser radar phase offset assignment and conflict detection

[0036]

[0037]

[0038] Further, the application also includes: a space occupation data calculation unit for calculating space occupation data based on the three-dimensional space model; a space occupation rate calculation unit for defining a reference space of a target area and calculating the space occupation rate in combination with the space occupation data.

[0039] Specifically, all objects that may hinder the passage are identified by the three-dimensional space model. The calculation process includes measuring the circumscribed cube size of each obstacle, counting the space volume occupied by each obstacle, and recording the specific location coordinates. For example, when a bicycle with a length of 8 meters, a width of 0.7 meters, and a height of 1.2 meters is detected, it is calculated that each obstacle occupies a space volume of 1.5 cubic meters. The reference space of the target area refers to the standard passage range of the fire passage defined in advance according to the building specifications and actual use requirements. For example, a space with a length of 10 meters, a width of 2 meters, and a height of 2.5 meters is taken as the reference, which needs to be kept completely unobstructed to ensure the evacuation requirements in emergency situations. The setting of the reference space needs to consider the characteristics of the building structure and the use scene, such as the hospital corridor may need a wider passage space than the ordinary office building. Comparing the obstacle volume with the total volume of the reference space, and then calculating the three-dimensional volume ratio, can more accurately reflect the actual passage obstruction degree. For example, when the total occupied volume reaches 3 cubic meters, and the reference space is 50 cubic meters, the occupancy rate is 6%, which ensures that the movement and changes of objects in the fire passage can be responded to in time.

[0040] Further, the present application further comprises: a contour feature extraction channel for distinguishing static environment and dynamic occupying objects from the three-dimensional space model by using a semantic segmentation algorithm, and extracting the contour features of the occupying objects according to the semantic segmentation result; a first space occupancy rate calculation channel for calculating the space occupancy rate based on the contour features.

[0041] Specifically, distinguishing static environment and dynamic objects in a three-dimensional space model using a semantic segmentation algorithm refers to inputting a three-dimensional space model reconstructed from point cloud data generated by a laser radar into a semantic analysis model, and classifying each point, thereby distinguishing fixed and immobile building structures, walls, ground, and other static backgrounds from temporary obstacles such as vehicles, cartons, and carts that may appear. The semantic segmentation algorithm is a deep learning technology that can identify and label points in space data, allowing rapid understanding of the category attributes of objects in the channel without relying on manual judgment. Extracting the contour features of the occupying object according to the semantic segmentation result refers to extracting the contour boundary from the point cloud belonging to the dynamic category after semantic classification is completed, and constructing the spatial shape information of the object. The contour features include not only the geometric dimensions such as length, width, and height of the object, but also the surface connectivity, edge sharpness, and coordinate distribution of the object in three-dimensional space. For example, when an object with a length of 1.2 meters, a width of 0.7 meters, and a height of 0.9 meters is detected, the volume boundary of the occupied area can be determined, laying the foundation for subsequent occupancy rate calculation. The contour extraction process can combine point cloud boundary detection algorithms, such as local description methods based on curvature changes or point density mutations. Calculating the spatial occupancy rate based on the contour features refers to calculating the volume ratio between the three-dimensional size of the object reflected by the contour features and the preset passable area of the channel, thereby obtaining the current occupied space ratio. The calculation of the spatial occupancy rate needs to define a standard reference space first, for example, a channel area with a length of 6 meters, a width of 2 meters, and a height of 2.5 meters corresponds to a total volume of 30 cubic meters. If the extracted occupying object volume is 1.35 cubic meters, the corresponding occupancy rate is 4.5%, which can dynamically reflect the degree of influence on the passage brought by different objects and different positions, and improve the judgment accuracy.

[0042] Further, the present application also includes: a second spatial occupancy rate calculation channel for dividing an effective passable area from a reference space of a target area, and calculating a spatial occupancy rate of the spatial occupancy data based on the effective passable area.

[0043] Specifically, the effective passing area is divided from the reference space of the target area, which means determining a complete space boundary as the reference space, and the reference space is set according to the building specifications, actual passing needs and scene purposes, for example, the length is 10 meters, the width is 2 meters, the height is 2.5 meters, and the total volume is 50 cubic meters. In order to more accurately evaluate the degree of obstruction of the fire passage, it is necessary to further demarcate the effective passing area actually used for the passing of people or vehicles from the reference space, and the effective passing area excludes the areas that cannot be passed, such as wall corners, edges or structural fixed equipment. For example, in a hospital corridor, in order to avoid wall fixed facilities, the actual effective passing area may be only a middle strip area with a width of 1.5 meters. The space occupancy rate of the space occupancy data calculated based on the effective passing area means that the passing range demarcated is taken as a reference framework, the space volume occupied by the dynamic object in the passing range is counted, and the volume is calculated by ratio with the volume of the effective passing area, so that whether there is a substantial obstruction to normal passing in the fire passage can be more accurately reflected. For example, if an occupying object such as a bicycle is located in the non-passable area at the edge of the passage, it will not be included in the occupancy rate; but if it blocks a passing space of 1.2 meters wide, 1.8 meters long and 1 meter high, it occupies an effective volume of about 2.16 cubic meters, and in an area with a passing volume of 30 cubic meters, the occupancy rate is 7.2%.

[0044] Further, the application also includes: a comparison unit for comparing the space occupancy rate with a preset dynamic threshold of space occupancy, and the preset dynamic threshold is adaptively adjusted according to environmental noise and point cloud density; an update triggering unit for triggering the update of the preset dynamic threshold when the space occupancy rate is greater than the preset dynamic threshold and the occupying duration is greater than the preset anti-transient interference time, and retaining the space occupancy rate.

[0045] Specifically, the space occupancy rate is compared with the preset dynamic threshold of space occupancy, which means that after the calculation of the volume of the obstacle in the effective passing area is completed, the space occupancy rate index is compared with the preset alarm threshold, so as to judge whether there is an abnormal occupation situation at present. The space occupancy rate represents the percentage of the occupied volume and the effective passing volume, which is a real-time changing index, and the preset dynamic threshold is a preset judgment basis, which can be adaptively adjusted with the change of the environment. Through the comparison mechanism, false responses to occasional object movement or short-time blocking can be avoided. The preset dynamic threshold is adaptively adjusted according to the environmental noise and the point cloud density, which means that a fixed threshold standard is not used in the actual operation process, but the preset dynamic threshold is intelligently adjusted according to the current monitoring environment. The environmental noise includes external interference factors that may cause abnormal point cloud data, such as light changes, wind blowing foreign matters, and rain and snow weather; the point cloud density refers to the number of laser points collected in a unit space, and the higher the density, the stronger the perception resolution. Through comprehensive dynamic calculation, the preset dynamic threshold can be set more sensitive under ideal conditions of high point cloud density and small environmental interference, and appropriately relaxed under harsh conditions to reduce the false alarm rate.

[0046] When the space occupancy rate is greater than the preset dynamic threshold and the occupation duration is greater than the preset anti-transient interference time, the update of the preset dynamic threshold is triggered, which means that after it is continuously monitored that the space occupancy rate exceeds the preset dynamic threshold and the duration of this state exceeds the preset anti-interference time window, it is considered that the occupation event is a real and persistent phenomenon rather than a short-time abnormality, and then the update mechanism of the preset dynamic threshold is started. The preset anti-transient interference time is usually set to a few seconds, for example, 3 seconds or 5 seconds, to exclude point cloud disturbance caused by temporary stay or rapid passing of personnel. The space occupancy rate is reserved, which means that when the threshold update is triggered, the current recorded occupancy rate value is reserved as key event data in the database for subsequent event analysis, risk level evaluation or multi-modal alarm decision. The data reservation mechanism can realize continuous tracking of the occupation event and provide a historical data basis for trend analysis and management strategy optimization.

[0047] Further, the application also includes: an occupation event acquisition channel for acquiring an occupation event when the space occupancy rate is greater than the preset dynamic threshold and the occupation duration is greater than the preset anti-transient interference time after the update; a reality confirmation channel for confirming the reality of the occupation event through a multi-device cooperative verification mechanism, and triggering an alarm according to the confirmation result.

[0048] Specifically, even if the preset dynamic threshold has been updated, if the new real-time monitoring data still shows that the space occupancy rate exceeds the current preset dynamic threshold, and the duration of this state still exceeds the set anti-interference time window, it will be formally judged that this case is a valid occupation event. The occupation event refers to the occurrence of a persistent obstacle behavior that hinders the normal passage of personnel or equipment in the fire passage or emergency area, such as piling up of sundries, parking of electric vehicles, or long-term temporary structures. The occupation duration is a time threshold set by the system, such as 5 seconds, 10 seconds, or longer, to distinguish between short-term passing and long-term hindering.

[0049] The multi-device cooperative verification mechanism confirms the authenticity of the occupation event, which means that after a preliminary judgment of an occupation event, an alarm is not triggered immediately, but the data of other sensor devices deployed in the same area are called to cross-verify, which can be multiple laser radars, cameras, millimeter wave radars, or infrared sensors. By fusing the perception information, the accuracy of the judgment is improved. The cooperative verification mechanism can effectively filter false positives caused by single device failure, obstruction, or occasional interference. For example, when a laser radar detects that the passage is occupied, but the nearby camera image does not show obvious obstacles, it can be considered that the result is uncertain and the alarm is temporarily suspended. According to the confirmation result, the alarm is triggered, which means that when multiple devices cross-verify consistently confirm that the occupation event does exist, the alarm action is executed. The alarm can be in various forms, including local sound and light warning, remote platform notification, SMS push, or linkage with the building fire fighting system for processing. By using the confirmation result as the alarm trigger condition, not only the credibility of the system's judgment is improved, but also frequent false alarms are avoided, which interferes with the normal work of the maintenance personnel.

[0050] Further, the present application also includes a multi-modal alarm layer for executing multi-modal alarm based on the occupation event, the multi-modal alarm including local alarm and cloud alarm.

[0051] Specifically, the multi-modal alarm based on the occupation event means that after confirming the occupation event, it will no longer rely on a single alarm method, but will trigger multiple types of alarm means at the same time or according to a strategy. The occupation event refers to the detection of persistent obstacles that hinder normal passage or safety requirements in passage channels, evacuation paths, or critical fire areas, such as temporary movement of shelves, piling up of sundries, parking of electric vehicles, etc. Multi-modal alarm refers to the use of multiple alarm methods according to the severity of the event, the level of the location, or the time factor to ensure that event information can be quickly and accurately transmitted to different response subjects.

[0052] The multi-modal alarm includes local alarm and cloud alarm. The local alarm can be directly activated by the local sound and light alarm device at the scene of the occupation event, and the event information can be uploaded to the remote management platform or the cloud server. The cloud system records, analyzes and forwards the event. The local alarm includes a buzzer, a warning light or a field broadcast loudspeaker, which immediately reminds the on-site personnel to quickly remove the occupation obstacle and restore the smooth passage. The cloud alarm uploads alarm information such as event time, space occupation rate value, channel number, and on-site image to the server for real-time viewing or historical backtracking by the background operation and maintenance personnel, and can be pushed to the on-duty personnel or safety person in charge through SMS, WeChat, email and other ways.

[0053] Further, the application further includes: a multi-modal alarm sub-layer, wherein the local alarm includes sound and light warning, and the cloud alarm includes a three-dimensional position of the occupation object and a material classification result based on reflection intensity.

[0054] Specifically, the local alarm includes sound and light warning, which means that when a valid occupation event is detected, a warning will be given to the surrounding personnel through the combination of sound and light, so as to achieve timely intervention. The sound warning usually uses a buzzer or a voice broadcast module, which can play a preset warning sound or voice prompt, such as “fire passage is occupied, please clean up immediately”; the light warning enhances attention through warning lights, flashing lights or projected signs, etc. in vision. The sound and light warning is a kind of active intervention means, which is suitable for environments that require quick guidance of personnel action, such as large shopping malls, hospital corridors or subway passages, which helps to promote the handling of illegal occupation behavior at the initial stage of the event.

[0055] The cloud alarm includes a three-dimensional position of the occupation object and a material classification result based on reflection intensity, which means that in the remote alarm information, not only the occurrence state of the event is uploaded, but also the accurate position coordinates of the occupation object in the three-dimensional space and the material type information of the object. The three-dimensional position refers to the spatial position of the occupation object on the X, Y and Z coordinate axes, which is used to accurately locate the specific point of the obstacle in the passage. The material classification based on reflection intensity refers to distinguishing the reflection characteristics of the object surface by analyzing the strength characteristics of the laser radar echo signal, so as to distinguish the materials such as metal, plastic and fabric, and further help the management system to judge the danger, mobility or category of the object, such as identifying metal shelves, wooden pallets or plastic buckets, so as to support different handling strategies.

[0056] In summary, the fire passage occupation identification system based on laser and radar cooperation provided by the application has the following technical effects: by achieving the technical target of high-precision occupation identification and intelligent alarm based on multi-source data fusion and dynamic threshold judgment, the technical effects of improving detection stability, reducing false alarm rate, enhancing system adaptability and linkage response capability are achieved.

[0057] Embodiment two, based on the same inventive concept as the fire passage occupation recognition system under laser and radar cooperation in the preceding embodiments, the present application also provides a fire passage occupation recognition method under laser and radar cooperation, please refer to the attached Figure 2 , comprising: emitting laser beams to the target area by solid-state laser radar, receiving reflected signals, extracting time difference and phase difference, converting to polar coordinate data, and generating point cloud data set; performing point cloud filtering, clustering and three-dimensional reconstruction based on the point cloud data set to construct a three-dimensional space model of the target area; calculating the space occupancy rate according to the three-dimensional space model, and when detecting that the space occupancy rate exceeds the preset dynamic threshold, performing adaptive occupation judgment and pushing alarm information.

[0058] Further, the fire passage occupation recognition method under laser and radar cooperation further comprises: assigning a unique phase offset to each solid-state laser radar in the solid-state laser radar, configuring the emission time of the laser beam according to the unique phase offset, and if a phase conflict is detected, reassigning a unique phase offset.

[0059] Further, the fire passage occupation recognition method under laser and radar cooperation further comprises: calculating space occupancy data based on the three-dimensional space model; defining a reference space of the target area, and calculating the space occupancy rate in combination with the space occupancy data.

[0060] Further, the fire passage occupation recognition method under laser and radar cooperation further comprises: using a semantic segmentation algorithm to distinguish static environment and dynamic occupation objects in the three-dimensional space model, and extracting the contour features of the occupation objects according to the semantic segmentation results; calculating the space occupancy rate based on the contour features.

[0061] Further, the fire passage occupation recognition method under laser and radar cooperation further comprises: dividing an effective passing area from the reference space of the target area, and calculating the space occupancy rate of the space occupancy data based on the effective passing area.

[0062] Further, the fire passage occupation recognition method under laser and radar cooperation further comprises: comparing the space occupancy rate with a preset dynamic threshold of space occupancy, and the preset dynamic threshold is adaptively adjusted according to environmental noise and point cloud density; when the space occupancy rate is greater than the preset dynamic threshold and the occupation duration is greater than the preset anti-transient interference time, triggering the update of the preset dynamic threshold, and retaining the space occupancy rate.

[0063] Further, the fire passage occupation recognition method under the cooperation of laser and radar further comprises: updating still satisfies the condition that the space occupation rate is greater than the preset dynamic threshold and the occupation duration is greater than the preset anti-transient interference time, obtaining an occupation event; verifying the authenticity of the occupation event through a multi-device cooperative verification mechanism, and triggering an alarm according to the confirmation result.

[0064] Further, the fire passage occupation recognition method under the cooperation of laser and radar further comprises: executing a multi-modal alarm based on the occupation event, and the multi-modal alarm comprises a local alarm and a cloud alarm.

[0065] Further, the fire passage occupation recognition method under the cooperation of laser and radar further comprises: the local alarm comprises an audible and visual warning, and the cloud alarm comprises a three-dimensional position of the occupied object and a material classification result based on the reflection intensity.

[0066] Each of the embodiments in the specification is described in a progressive manner, and each embodiment focuses on the difference from other embodiments. The laser and radar cooperative fire passage occupation recognition system in the first embodiment and the specific examples are also applicable to the laser and radar cooperative fire passage occupation recognition method of the present embodiment. Through the foregoing detailed description of the laser and radar cooperative fire passage occupation recognition system, those skilled in the art can clearly understand the laser and radar cooperative fire passage occupation recognition method in the present embodiment. Therefore, in order to make the specification concise, it will not be described in detail here.

[0067] The above description of the disclosed embodiments enables a person skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0068] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application also intends to include these modifications and variations.

Claims

1. The fire passage occupation identification system based on the cooperation of laser and radar is characterized by: include: The point cloud generation module is used to transmit a laser beam to the target area through a solid-state laser radar, receive the reflected signal, extract the time difference and phase difference, convert them into polar coordinate data, and generate a point cloud data set; A three-dimensional space model construction module is used to perform point cloud filtering, clustering and three-dimensional reconstruction based on the point cloud data set to construct a three-dimensional space model of the target area; The occupancy judgment module is used to calculate the space occupancy rate according to the three-dimensional space model, and when it is detected that the space occupancy rate exceeds a preset dynamic threshold, perform adaptive occupancy judgment and push alarm information.

2. The fire passage occupancy identification system based on the cooperation of laser and radar as claimed in claim 1 is characterized in that: The point cloud generation module includes: a radar coordination unit, which is used to assign a unique phase offset to each solid-state laser radar in the solid-state laser radar, configure the emission time of the laser beam according to the unique phase offset, and reallocate the unique phase offset if a phase conflict is detected.

3. The fire passage occupancy identification system based on laser and radar collaboration as claimed in claim 1 is characterized in that: The occupancy determination module includes: a space occupancy data calculation unit, configured to calculate space occupancy data based on the three-dimensional space model; The space occupancy rate calculation unit is used to define a reference space of the target area and calculate the space occupancy rate based on the space occupancy data.

4. The fire passage occupancy identification system based on the cooperation of laser and radar as claimed in claim 3 is characterized in that: The space occupancy rate calculation unit includes: a contour feature extraction channel for distinguishing static environments from dynamic occupied objects in the three-dimensional space model using a semantic segmentation algorithm, and extracting contour features of the occupied objects based on the semantic segmentation results; The first space occupancy calculation channel is used to calculate the space occupancy based on the contour feature.

5. The fire passage occupation identification system based on the cooperation of laser and radar as claimed in claim 3 is characterized in that: The space occupancy rate calculation unit further includes: a second space occupancy rate calculation channel, which is used to divide an effective passage area from a reference space of the target area, and calculate the space occupancy rate of the space occupancy data based on the effective passage area.

6. The fire passage occupancy identification system based on the cooperation of laser and radar as claimed in claim 1 is characterized in that: The occupancy determination module further includes: a comparing unit, configured to compare the space occupancy rate with a preset dynamic threshold of space occupancy, wherein the preset dynamic threshold is adaptively adjusted according to environmental noise and point cloud density; An update triggering unit is used to trigger the update of the preset dynamic threshold and retain the space occupancy when the space occupancy is greater than the preset dynamic threshold and the occupancy duration is greater than the preset anti-instantaneous interference time.

7. The fire passage occupation identification system based on the cooperation of laser and radar as claimed in claim 6 is characterized in that: The update triggering unit includes: An occupancy event acquisition channel, used to acquire an occupancy event when the conditions that the space occupancy rate is greater than the preset dynamic threshold and the occupancy duration is greater than the preset anti-instantaneous interference time are still met after the update; The authenticity confirmation channel is used to confirm the authenticity of the occupation event through a multi-device collaborative verification mechanism and trigger an alarm based on the confirmation result.

8. The fire passage occupation identification system based on the cooperation of laser and radar as claimed in claim 7 is characterized in that: The authenticity confirmation channel includes: a multimodal alarm layer, which is used to execute a multimodal alarm based on the occupation event, and the multimodal alarm includes a local alarm and a cloud alarm.

9. The fire passage occupation identification system based on the cooperation of laser and radar as claimed in claim 8, characterized in that: The multimodal alarm layer includes: a multimodal alarm sublayer, which is used for the local alarm to include sound and light warnings, and the cloud alarm to include the three-dimensional position of the occupied object and the material classification result based on reflection intensity.

10. A method for identifying fire passage occupation using laser and radar collaboration, characterized in that: The method is implemented by a fire passage occupancy identification system coordinated by laser and radar according to any one of claims 1 to 9, comprising: The solid-state laser radar emits a laser beam to the target area, receives the reflected signal, extracts the time difference and phase difference, converts them into polar coordinate data, and generates a point cloud data set; Performing point cloud filtering, clustering, and three-dimensional reconstruction based on the point cloud dataset to construct a three-dimensional spatial model of the target area; The space occupancy rate is calculated according to the three-dimensional space model. When it is detected that the space occupancy rate exceeds a preset dynamic threshold, an adaptive occupancy decision is performed and an alarm message is pushed.

Images