Smart park risk warning method, system, equipment and medium based on 3D modeling

Abstract

The present application provides a risk warning method, system, equipment and medium for a smart park based on 3D modeling, which obtains building information and road information in the smart park; performs 3D modeling based on component information and topological relationships to obtain 3D models of buildings and parks; divides the 3D model of the park into risk and non-risk areas; performs modeling based on non-risk area images to obtain a 3D model of personnel, and predicts the 3D model of personnel at the target time to obtain a predicted 3D model; determines whether each predicted 3D model appears in the risk area based on the model integration result; if so, performs a risk warning for the park through a park alarm device; authenticates the target non-risk personnel at the target time, and issues a park risk alarm through a remote alarm device. The above method is used to improve the effectiveness and stability of risk warnings for smart parks, while improving the response speed and efficiency to risk warnings.

Description

Technical Field

[0001] The present invention relates to the field of smart parks, and in particular to a smart park risk warning method, system, equipment and medium based on 3D modeling. Background Art

[0002] At present, in the field of park management, as the importance of park safety management continues to increase, the possibilities and types of emergencies faced by parks during operation are becoming increasingly complex and diverse. In order to improve the safety of park operations, some areas in the park that may have usage risks, or equipment and areas that undertake emergency rescue tasks, are usually restricted in response to possible emergencies, so as to achieve effective management of park safety.

[0003] In the prior art, when restricting the use of risk areas in the park, several park inspectors are usually set up to patrol the park. When the inspectors find that someone has the intention to enter the risk area or use emergency rescue equipment without authorization, they will step forward to warn or stop them, so as to achieve early warning of park risks; or when it is found that someone has entered the risk area or used emergency rescue equipment without authorization, the risk alarm is manually triggered. However, it is found in the study that when analyzing and judging the intention of people to enter the risk area or use emergency rescue equipment in the above way, it depends on the subjective analysis and experience judgment of the inspectors, which is easily affected by subjective factors, increasing the uncertainty of decision-making, thereby reducing the effectiveness and stability of risk warning for smart parks. In addition, since the risk alarm needs to be manually triggered by the inspectors when it is found that someone has entered the risk area or used emergency rescue equipment without authorization, and the manual triggering of the risk alarm usually requires multi-level manual approval and communication, and requires cross-departmental and cross-regional collaborative processing, which will lead to problems such as slow response speed and efficiency of risk warning. Summary of the invention

[0004] In view of this, the purpose of the present invention is to provide a smart park risk warning method, system, equipment and medium based on 3D modeling, so as to improve the effectiveness and stability of risk warning for smart parks, and at the same time improve the response speed and efficiency to risk warning.

[0005] In a first aspect, an embodiment of the present application provides a smart park risk warning method based on 3D modeling, which is applied to a smart park risk warning system based on 3D modeling. The system includes an information collection module, a 3D modeling module, a risk assessment module, and a risk alarm module. The risk alarm module includes a park alarm device and a remote alarm device. The method includes:

[0006] The information collection module obtains the building information of each building in the smart park and the road information of each road in the smart park, wherein the building information includes the component information of each building component and the first topological relationship between each building component, and the road information includes the road type of each road and the second topological relationship between each road and each building;

[0007] The 3D modeling module performs 3D modeling on each building based on the component information of each building component in each building and the first topological relationship between each building component to obtain a 3D building model of each building, and performs 3D modeling on the smart park based on the 3D building model of each building, the road type of each road and the second topological relationship between each road and each building to obtain a 3D park model of the smart park;

[0008] The information acquisition module divides the 3D model of the smart park into risk areas and non-risk areas according to the component information of each building component and the road type of each road, and obtains the risk area image of each risk area and the non-risk area image of each non-risk area in real time;

[0009] The 3D modeling module performs 3D modeling on non-risk personnel located in each non-risk area based on each non-risk area image to obtain a 3D model of each non-risk person, and predicts the 3D model of each non-risk person at a target time based on the 3D model of each non-risk person at different times to obtain a predicted 3D model of each non-risk person;

[0010] The risk assessment module integrates the predicted 3D models of each non-risk person with the park 3D model, and determines whether the predicted 3D models of each non-risk person appear in the risk area of ​​the park 3D model according to the integration result;

[0011] If there is a target non-risk person whose 3D model is predicted to appear in the risk area of ​​the park 3D model, the risk alarm module issues a park risk warning for the target non-risk person and the risk area through the park alarm device;

[0012] At the target moment, the risk assessment module determines the identity information of the risk persons in each risk area based on the risk area images of each risk area, and performs identity authentication on the target non-risk person according to the identity information of each risk person, and determines whether the target non-risk person is one of the risk persons according to the identity authentication result;

[0013] If the target non-risk person is one of the risk persons, the risk alarm module will issue a park risk alarm for the target non-risk person and the risk area through the remote alarm device.

[0014] Optionally, performing 3D modeling on each building based on the component information of each building component in each building and the first topological relationship between each building component to obtain a 3D building model of each building includes:

[0015] For each building, the coordinates of each building component and the connected space in the building are determined according to the component information of each building component in the building and the first topological relationship between the building components;

[0016] An indoor road network model of the building is constructed according to the coordinate components of each building component and the connected space in the building;

[0017] A 3D building model of the building is constructed according to the indoor road network model of the building, the scaling ratio of the indoor road network model and the types of each building component.

[0018] Optionally, the performing 3D modeling of the smart park based on the 3D building models of the buildings, the road types of the roads, and the second topological relationship between the roads and the buildings to obtain the 3D park model of the smart park includes:

[0019] Adjusting the scaling ratio of the 3D building models of the various buildings so that the 3D building models of the various buildings have the same target scaling ratio;

[0020] Generate an outdoor road network model of the smart park according to the road type of each road and the second topological relationship between each road and each building;

[0021] A 3D model of the smart park is constructed based on the 3D building models of each building, the outdoor road network model and the target scaling ratio.

[0022] Optionally, the component information includes a component type, the component type is emergency equipment or non-emergency equipment, and the road type is a safe passage or a non-safe passage; and the 3D model of the smart park is divided into risk areas and non-risk areas according to the component information of each building component and the road type of each road, including:

[0023] Determine whether the component type of each building component is an emergency device, and determine whether the road type of each road is a safe passage;

[0024] Taking the location of each emergency building component as the center point, regional expansion is performed to obtain a number of risk sub-regions, and regional expansion is performed on each emergency road to obtain a number of risk sub-roads, wherein the emergency building component is a building component whose component type is emergency equipment, and the emergency road is a road whose road type is a safe passage;

[0025] Each risk sub-area and each risk sub-road is marked in the park 3D model to obtain the risk area, and the area other than the risk area in the park 3D model is marked as the non-risk area.

[0026] Optionally, the performing 3D modeling of the non-risk personnel located in each non-risk area based on each non-risk area image to obtain a 3D model of each non-risk personnel includes:

[0027] For each non-risk area image, the static feature data of the non-risk area image is extracted using a feature extraction model to obtain human feature data and background feature data of the non-risk area image;

[0028] Generate a human sub-image and a background sub-image of the non-risk area image based on the human feature data and background feature data of the non-risk area image respectively;

[0029] According to the image boundary pixel value of the background sub-image of the non-risk area image, the image boundary of the human body sub-image of the non-risk area image is smoothed to obtain the target human body sub-image of the non-risk area image;

[0030] Performing human body key point recognition on each target human body sub-image to obtain human body modeling data of each target human body sub-image;

[0031] 3D modeling is performed based on the human body modeling data of each target human body sub-image to obtain a 3D model of each non-risk person corresponding to each target human body sub-image.

[0032] Optionally, predicting the personnel 3D model of each non-risk person at the target time based on the personnel 3D model of each non-risk person at different times to obtain the predicted 3D model of each non-risk person includes:

[0033] For each risk person, the current action and action change information of the non-risk person are determined based on the 3D model of the non-risk person at different times;

[0034] Determine the predicted action of the non-risk person at the target moment according to the current action and action change information of the non-risk person;

[0035] A predicted 3D model of the non-risk person is constructed according to the predicted actions of the non-risk person at the target moment.

[0036] Optionally, determining the identity information of the risk personnel in each risk area based on the risk area image of each risk area, and performing identity authentication on the target non-risk personnel according to the identity information of each risk personnel, includes:

[0037] For each risk area, face recognition is performed on the risk area image of the risk area to determine the face feature value of the recognized face, and the face feature value of the recognized face is used as the identity information of the risk person corresponding to the face in the risk area;

[0038] Obtain the facial feature value of the target non-risk person, determine whether the facial feature value of each risk person is the same as the facial feature value of the target non-risk person, and authenticate the identity of the target non-risk person based on the determination result.

[0039] In a second aspect, an embodiment of the present application provides a smart park risk warning system based on 3D modeling, the system comprising an information collection module, a 3D modeling module, a risk assessment module and a risk alarm module, the risk alarm module comprising a park alarm device and a remote alarm device;

[0040] The information collection module is used to obtain building information of each building in the smart park and road information of each road in the smart park, wherein the building information includes component information of each building component and a first topological relationship between each building component, and the road information includes a road type of each road and a second topological relationship between each road and each building;

[0041] The 3D modeling module is used to perform 3D modeling on each building based on the component information of each building component in each building and the first topological relationship between each building component to obtain a 3D building model of each building, and perform 3D modeling on the smart park based on the 3D building model of each building, the road type of each road and the second topological relationship between each road and each building to obtain a 3D park model of the smart park;

[0042] The information acquisition module is used to divide the 3D model of the smart park into risk areas and non-risk areas according to the component information of each building component and the road type of each road, and obtain the risk area image of each risk area and the non-risk area image of each non-risk area in real time;

[0043] The 3D modeling module is used to perform 3D modeling on non-risk personnel located in each non-risk area based on each non-risk area image to obtain a 3D model of each non-risk person, and predict the 3D model of each non-risk person at a target time based on the 3D model of each non-risk person at different times to obtain a predicted 3D model of each non-risk person;

[0044] The risk assessment module is used to integrate the predicted 3D models of each non-risk person with the park 3D model, and determine whether the predicted 3D model of each non-risk person appears in the risk area of ​​the park 3D model according to the integration result;

[0045] The risk alarm module is used to issue a park risk warning for the target non-risk person and the risk area through the park alarm device if there is a target non-risk person whose 3D model is predicted to appear in the risk area of ​​the park 3D model;

[0046] The risk assessment module is used to determine the identity information of the risk persons in each risk area based on the risk area images of each risk area at the target time, and to authenticate the target non-risk person according to the identity information of each risk person, and to determine whether the target non-risk person is one of the risk persons according to the authentication result;

[0047] If the target non-risk person is one of the risk persons, the risk alarm module will issue a park risk alarm for the target non-risk person and the risk area through the remote alarm device.

[0048] Optionally, performing 3D modeling on each building based on the component information of each building component in each building and the first topological relationship between each building component to obtain a 3D building model of each building includes:

[0049] For each building, the coordinates of each building component and the connected space in the building are determined according to the component information of each building component in the building and the first topological relationship between the building components;

[0050] An indoor road network model of the building is constructed according to the coordinate components of each building component and the connected space in the building;

[0051] A 3D building model of the building is constructed according to the indoor road network model of the building, the scaling ratio of the indoor road network model and the types of each building component.

[0052] Optionally, the performing 3D modeling of the smart park based on the 3D building models of the buildings, the road types of the roads, and the second topological relationship between the roads and the buildings to obtain the 3D park model of the smart park includes:

[0053] Adjusting the scaling ratio of the 3D building models of the various buildings so that the 3D building models of the various buildings have the same target scaling ratio;

[0054] Generate an outdoor road network model of the smart park according to the road type of each road and the second topological relationship between each road and each building;

[0055] A 3D model of the smart park is constructed based on the 3D building models of each building, the outdoor road network model and the target scaling ratio.

[0056] Optionally, the component information includes a component type, the component type is emergency equipment or non-emergency equipment, and the road type is a safe passage or a non-safe passage; and the 3D model of the smart park is divided into risk areas and non-risk areas according to the component information of each building component and the road type of each road, including:

[0057] Determine whether the component type of each building component is an emergency device, and determine whether the road type of each road is a safe passage;

[0058] Taking the location of each emergency building component as the center point, regional expansion is performed to obtain a number of risk sub-regions, and regional expansion is performed on each emergency road to obtain a number of risk sub-roads, wherein the emergency building component is a building component whose component type is emergency equipment, and the emergency road is a road whose road type is a safe passage;

[0059] Each risk sub-area and each risk sub-road is marked in the park 3D model to obtain the risk area, and the area other than the risk area in the park 3D model is marked as the non-risk area.

[0060] Optionally, the performing 3D modeling of the non-risk personnel located in each non-risk area based on each non-risk area image to obtain a 3D model of each non-risk personnel includes:

[0061] For each non-risk area image, the static feature data of the non-risk area image is extracted using a feature extraction model to obtain human feature data and background feature data of the non-risk area image;

[0062] Generate a human sub-image and a background sub-image of the non-risk area image based on the human feature data and background feature data of the non-risk area image respectively;

[0063] According to the image boundary pixel value of the background sub-image of the non-risk area image, the image boundary of the human body sub-image of the non-risk area image is smoothed to obtain the target human body sub-image of the non-risk area image;

[0064] Performing human body key point recognition on each target human body sub-image to obtain human body modeling data of each target human body sub-image;

[0065] 3D modeling is performed based on the human body modeling data of each target human body sub-image to obtain a 3D model of each non-risk person corresponding to each target human body sub-image.

[0066] Optionally, predicting the personnel 3D model of each non-risk person at the target time based on the personnel 3D model of each non-risk person at different times to obtain the predicted 3D model of each non-risk person includes:

[0067] For each risk person, the current action and action change information of the non-risk person are determined based on the 3D model of the non-risk person at different times;

[0068] Determine the predicted action of the non-risk person at the target moment according to the current action and action change information of the non-risk person;

[0069] A predicted 3D model of the non-risk person is constructed according to the predicted actions of the non-risk person at the target moment.

[0070] Optionally, determining the identity information of the risk personnel in each risk area based on the risk area image of each risk area, and performing identity authentication on the target non-risk personnel according to the identity information of each risk personnel, includes:

[0071] For each risk area, face recognition is performed on the risk area image of the risk area to determine the face feature value of the recognized face, and the face feature value of the recognized face is used as the identity information of the risk person corresponding to the face in the risk area;

[0072] Obtain the facial feature value of the target non-risk person, determine whether the facial feature value of each risk person is the same as the facial feature value of the target non-risk person, and authenticate the identity of the target non-risk person based on the determination result.

[0073] In the third aspect, an embodiment of the present application provides a computer device, comprising: a processor, a memory and a bus, the memory storing machine-readable instructions executable by the processor, and when the computer device is running, the processor and the memory communicate through the bus, and when the machine-readable instructions are executed by the processor, the steps of the smart park risk warning method based on 3D modeling described in any optional implementation manner of the second aspect are performed.

[0074] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the smart park risk warning method based on 3D modeling described in any optional implementation manner in the second aspect above are executed.

[0075] The technical solution provided by this application includes but is not limited to the following beneficial effects:

[0076] First, the present application obtains the building information of each building in the smart park and the road information of each road in the smart park through the information collection module, and the 3D modeling module performs 3D modeling on each building based on the component information of each building component in each building and the first topological relationship between each building component to obtain the building 3D model of each building, and performs 3D modeling on the smart park based on the building 3D model of each building, the road type of each road and the second topological relationship between each road and each building to obtain the park 3D model of the smart park. 3D modeling can be performed on the smart park to obtain the park 3D model, and the structural shape composition of the smart park can be abstracted by computer to obtain a quantitative modeling result, which is convenient for the subsequent quantitative analysis and prediction of risk behaviors in the smart park based on the quantitative modeling results.

[0077] Then, the information acquisition module divides the 3D model of the smart park into risk areas and non-risk areas according to the component information of each building component and the road type of each road, and obtains the risk area image of each risk area and the non-risk area image of each non-risk area in real time; the 3D modeling module performs 3D modeling on the non-risk personnel located in each non-risk area based on the non-risk area images to obtain the 3D model of each non-risk personnel, and predicts the 3D model of each non-risk personnel at the target time based on the 3D model of each non-risk personnel at different times to obtain the predicted 3D model of each non-risk personnel; the risk assessment module integrates the predicted 3D model of each non-risk personnel with the park 3D model, and judges whether the predicted 3D model of each non-risk personnel appears in the risk area of ​​the park 3D model according to the integration result; through the above steps, 3D modeling is performed for the personnel located in the non-risk area, and quantitative analysis is performed to predict whether they have a trend of entering the risk area, so as to avoid the uncertainty and instability that may be caused by the analysis and judgment of human experience, and improve the accuracy of personnel behavior prediction.

[0078] Next, if there is a target non-risk person whose 3D model is predicted to appear in the risk area of ​​the park's 3D model, the risk alarm module will issue a park risk warning for the target non-risk person and the risk area through the park alarm device. This can enable the park risk warning to be issued in advance through the park alarm device when there is a potential park risk, thereby preventing and avoiding the occurrence of risk events as much as possible in advance and ensuring the safe operation of the smart park.

[0079] Finally, at the target moment, the risk assessment module determines the identity information of the risk personnel in each risk area based on the risk area image of each risk area, and authenticates the target non-risk personnel according to the identity information of each risk personnel, and determines whether the target non-risk personnel is one of the risk personnel according to the identity authentication result; if the target non-risk personnel is one of the risk personnel, the risk alarm module will issue a park risk alarm for the target non-risk personnel and the risk area through a remote alarm device; through the above steps, the park risk alarm is issued through a remote alarm device for risk events that cannot be avoided in advance according to the personnel identity authentication results, which can achieve timely synchronization of the occurrence of on-site events to the remote management end, improve the response speed and efficiency of risk alarms, and further improve the timeliness and effectiveness of park risk management.

[0080] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0082] Figure 1 A flowchart of a smart park risk warning based on 3D modeling provided in the first embodiment of the present invention is shown;

[0083] Figure 2 A flow chart of a method for generating a 3D model of a building provided in the first embodiment of the present invention is shown;

[0084] Figure 3 A flow chart of a method for generating a 3D model of a park provided in the first embodiment of the present invention is shown;

[0085] Figure 4 A flow chart of a method for dividing a region provided in Embodiment 1 of the present invention is shown;

[0086] Figure 5 A flow chart of a method for generating a 3D model of a person provided in the first embodiment of the present invention is shown;

[0087] Figure 6 A flowchart of a method for generating a predictive 3D model provided in the first embodiment of the present invention is shown;

[0088] Figure 7A flow chart of an identity authentication method provided by Embodiment 1 of the present invention is shown;

[0089] Figure 8 A schematic diagram of the structure of a smart park risk warning system based on 3D modeling provided in the second embodiment of the present invention is shown;

[0090] Fig. 9 A schematic diagram of the structure of a computer device provided in Embodiment 3 of the present invention is shown. DETAILED DESCRIPTION

[0091] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present invention.

[0092] Embodiment 1

[0093] To facilitate understanding of this application, Figure 1 The flowchart of a smart park risk warning method based on 3D modeling provided in the first embodiment of the present invention is described in detail to explain the first embodiment of the present application.

[0094] See also Figure 1 As shown, Figure 1 A flowchart of a smart park risk warning based on 3D modeling provided by an embodiment of the present invention is shown, wherein the flowchart is applied to a smart park risk warning system based on 3D modeling, the system includes an information collection module, a 3D modeling module, a risk assessment module and a risk alarm module, the risk alarm module includes a park alarm device and a remote alarm device, and the method includes steps S101 to S108:

[0095] S101: The information acquisition module obtains the building information of each building in the smart park and the road information of each road in the smart park, wherein the building information includes the component information of each building component and the first topological relationship between each building component, and the road information includes the road type of each road and the second topological relationship between each road and each building.

[0096] Specifically, component information includes component name, component type, component shape, component size, etc., and road information also includes road name, road shape, road size, etc. Component types mainly include columns, walls, doors, windows, stairs, elevators, corridors, and rooms. Topological information refers to the connection relationship between various components or things, mainly the connection relationship between vertical channels, floors, and rooms. Vertical channels include stairs and elevators.

[0097] S102: The 3D modeling module performs 3D modeling on each building based on the component information of each building component in each building and the first topological relationship between each building component to obtain a 3D building model of each building, and performs 3D modeling on the smart park based on the 3D building model of each building, the road type of each road and the second topological relationship between each road and each building to obtain a 3D park model of the smart park.

[0098] Specifically, the component information of each building component can describe the component position of a single component, and the first topological relationship between each building component can describe the relative position between each component. For each building, the component position of the central component in the building is obtained, and the positions of other components relative to the central component are determined based on the component position of the central component and the first topological relationship between each building component. A 3D model of the building is generated based on the component position of the central component and the positions of other components relative to the central component, and each node in the 3D model of the building is represented by the component information (component name) of each component.

[0099] The smart park is composed of several buildings, and the buildings are connected by roads. After obtaining the 3D models of the buildings, the roads between the buildings are also 3D modeled according to the modeling standard of the 3D models of the buildings, and a 3D model of the park including the 3D models of the buildings and the 3D models of the roads is obtained.

[0100] S103: The information acquisition module divides the 3D model of the smart park into risk areas and non-risk areas according to the component information of each building component and the road type of each road, and obtains the risk area image of each risk area and the non-risk area image of each non-risk area in real time.

[0101] Specifically, for parks (including communities, campuses, factories, etc.), safety passages and emergency equipment are usually set up. Safety passages are emergency walking and evacuation passages set up to ensure operational safety, and are important measures to ensure personnel safety. Safety passages can be divided into firewall safety passages and fire safety passages. Firewall safety passages are mainly used in commercial buildings to solve the problem of long evacuation distances and the first floor of the evacuation stairwell not directly leading to the outdoors. Fire safety passages are passages for firefighters to rescue and evacuate trapped people when accidents occur. They are usually set up in high-rise buildings to ensure that people can be evacuated quickly in emergency situations such as fires.

[0102] Emergency equipment refers to special equipment and instruments used to protect life, property and environment in emergencies or emergencies. Emergency equipment includes but is not limited to the following types. Life-saving equipment: such as life jackets, lifebuoys, first aid kits, etc., used to protect and save lives. Fire-fighting equipment: such as fire extinguishers, fire pumps, fire motorcycles, fire trucks, etc., used to extinguish fires and prevent the spread of fires. Emergency support-rescue fire truck recommendation. Communication equipment: such as walkie-talkies, satellite phones, digital cluster handheld terminals, etc., used to maintain contact and coordinate rescue work in emergencies. Detection equipment: such as infrared thermal imagers, water depth detectors, gas detectors, etc., used to detect and evaluate dangerous factors in the environment. Rescue equipment: such as rescue troop carriers, rope rescue tools, demolition tools, etc., used to rescue personnel and transport materials at the disaster site. Protective equipment: such as protective clothing, gas masks, protective helmets, etc., used to protect rescue personnel from harm.

[0103] In daily life, when no emergency occurs, safety passages are not allowed to be passed or occupied, and emergency equipment is not allowed to be used. In order to monitor the abnormal use of safety passages and emergency equipment, the 3D model of the smart park is divided into regions according to the component information of each building component and the road type of each road. The areas near the safety passages and emergency equipment are divided into risk areas, and other areas are divided into non-risk areas. The regional images of risk areas and non-risk areas are collected in real time. Based on the collected risk area images and non-risk area images of each non-risk area, it is determined whether there may be abnormal use of safety passages and emergency equipment.

[0104] S104: The 3D modeling module performs 3D modeling on the non-risk personnel located in each non-risk area based on the images of each non-risk area to obtain a 3D model of each non-risk personnel, and predicts the 3D model of each non-risk personnel at a target time based on the 3D model of each non-risk personnel at different times to obtain a predicted 3D model of each non-risk personnel.

[0105] Specifically, the personnel located in each non-risk area are initially marked as non-risk personnel, and the 3D modeling module performs 3D modeling on each non-risk personnel based on the non-risk area containing each non-risk personnel to obtain a 3D model of each non-risk personnel. The 3D model of personnel obtained at this time is a 3D model indicating the personnel information at the current moment. However, since the behavior of the personnel in the park is uncontrolled, the personnel originally in the risk area may enter the non-risk area at the next moment, or the personnel originally in the non-risk area may enter the risk area at the next moment. Therefore, in order to warn in advance of the situation where personnel appear in the risk area and predict whether the personnel will enter the risk area at the target moment, the 3D model of each non-risk personnel at the target moment is predicted based on the 3D model of each non-risk personnel at different moments, and the predicted 3D model of each non-risk personnel is obtained, and the 3D model indicating the personnel information at the target moment is obtained.

[0106] S105: The risk assessment module integrates the predicted 3D models of each non-risk person with the park 3D model, and determines whether the predicted 3D models of each non-risk person appear in the risk area of ​​the park 3D model according to the integration result.

[0107] Specifically, for non-risk personnel, before the target time, the position of their 3D model in the park 3D model is located in the non-risk area. If the non-risk personnel may move to the risk area at the target time, then after predicting the 3D model of the non-risk personnel at the target time by the above method, the position of the predicted 3D model in the park 3D model should be located in the risk area. Then, it can be judged whether each non-risk personnel is likely to appear in the risk area at the target time based on the position of the predicted 3D model of each non-risk personnel in the park 3D model.

[0108] S106: If there is a target non-risk person whose 3D model is predicted to appear in the risk area of ​​the park 3D model, the risk alarm module issues a park risk warning for the target non-risk person and the risk area through the park alarm device.

[0109] Specifically, if there is a target non-risk person whose predicted 3D model appears in the risk area of ​​the park 3D model, it means that the target non-risk person corresponding to the predicted 3D model may move into the risk area at the target time, and then a park risk warning is performed for the target non-risk person and the risk area through the park alarm device. The park alarm device is an alarm device set in the park, which includes a sound alarm device and a light alarm device. The park risk warning is performed through the park alarm device. When the target non-risk person has not entered the risk area yet, but has the tendency and possibility to enter the risk area, the target non-risk person and the risk area can be warned on the spot. Not only can the target non-risk person be warned to prevent him from entering the risk area, but other personnel in the park can also be informed of the occurrence of this situation, so that other personnel can prevent the target non-risk person from entering the risk area in time on the spot, and avoid the target non-risk person from entering the risk area.

[0110] If there is no target non-risk person whose predicted 3D model appears in the risk area of ​​the park 3D model, it means that the target non-risk person corresponding to the predicted 3D model cannot move into the risk area at the target time, and the park risk warning will not be issued through the park alarm device.

[0111] S107: At the target moment, the risk assessment module determines the identity information of the risk persons in each risk area based on the risk area images of each risk area, and authenticates the target non-risk person according to the identity information of each risk person, and determines whether the target non-risk person is one of the risk persons according to the identity authentication result.

[0112] Specifically, the warning before this step is a prediction and warning for possible abnormal occupation of risk areas, and the warning after this step is an analysis and alarm for abnormal occupation of risk areas that have already occurred.

[0113] First, the identity information of the target non-risk personnel is collected. At the target moment, the risk assessment module determines the identity information of the risk personnel in each risk area based on the risk area images of each risk area, and authenticates the target non-risk personnel based on the identity information of each risk personnel to determine whether the identity information of the target non-risk personnel is the same as the identity information of any risk personnel.

[0114] S108: If the target non-risk person is one of the risk persons, the risk alarm module issues a park risk alarm for the target non-risk person and the risk area through the remote alarm device.

[0115] Specifically, if the identity information of the target non-risk person is the same as the identity information of any risk person, it means that the target non-risk person is one of the risk persons, indicating that the target non-risk person has indeed entered the risk area from the non-risk area at the target time, and the abnormal occupation of the risk area has occurred at the target time. At this time, it means that the on-site alarm of the park alarm device is ineffective for the target non-risk person. The target non-risk person still enters the risk area under the alarm prompt of the park alarm device, so it is necessary to use the remote alarm device to perform a park risk alarm to remind the park management personnel that the abnormal occupation of the risk area has occurred. The remote alarm device includes a park management platform and a management terminal. The park management platform can be set in the park management room, and the management terminal can be carried by the park management personnel. When the park risk alarm is performed through the remote alarm device, the park management platform and the management terminal can respectively display the images of the target non-risk person and the risk area to the park management personnel, and issue an alarm prompt, so that the park management personnel can manage and intervene in the target non-risk person and the risk area according to the image information and the alarm prompt.

[0116] If the target non-risk person is not one of the risk persons, it means that the target non-risk person did not enter the risk area from the non-risk area at the target time, and there was no abnormal occupation of the risk area at the target time, so there is no need to issue a park risk alarm through the remote alarm device.

[0117] In one possible embodiment, see Figure 2 As shown, Figure 2 A flow chart of a method for generating a 3D model of a building provided in the first embodiment of the present invention is shown, wherein the 3D modeling of each building is performed based on the component information of each building component in each building and the first topological relationship between each building component to obtain the 3D model of each building, including steps S201-S203:

[0118] S201: For each building, determine the coordinates of each building component and the connected space in the building according to the component information of each building component in the building and the first topological relationship between the building components.

[0119] Specifically, the component position coordinates and geometric shapes of each building component are determined according to the component information of each building component, and the position coordinates and geometric shapes of all connected spaces in the target building can be determined according to the first topological relationship between the building components.

[0120] S202: Build an indoor road network model of the building according to the coordinates of the building components and the connected spaces in the building.

[0121] Specifically, each building component is abstracted as a node in the indoor road network model, each connected space is abstracted as a path in the indoor road network model, the starting point and end point of each path are determined according to the position coordinates and geometric shapes of each connected space, and the nodes connected by the starting point and end point of each path are determined according to the component position coordinates and geometric shapes of each building component. Each node and the paths between nodes are connected into a network to obtain the indoor road network model of the building.

[0122] S203: Constructing a 3D building model of the building according to the indoor road network model of the building, the scaling ratio of the indoor road network model and the types of each building component.

[0123] Specifically, the indoor road network model of the building is scaled according to the scaling ratio of the indoor road network model, and the type of building components indicated by each node in the indoor road network model is marked to obtain the building 3D model of the building.

[0124] In one possible embodiment, see Figure 3 As shown, Figure 3 A flow chart of a method for generating a 3D model of a park provided in the first embodiment of the present invention is shown, wherein the 3D modeling of the smart park is performed based on the 3D model of each building, the road type of each road, and the second topological relationship between each road and each building to obtain the 3D model of the park of the smart park, including steps S301-S303:

[0125] S301: adjusting the scaling ratio of the 3D building model of each building so that the 3D building model of each building has the same target scaling ratio.

[0126] S302: Generate an outdoor road network model of the smart park according to the road type of each road and the second topological relationship between each road and each building.

[0127] S303: Construct a 3D model of the smart park according to the 3D building models of each building, the outdoor road network model and the target scaling ratio.

[0128] Specifically, each building is abstracted as a node in the outdoor road network model, each road is abstracted as a path in the outdoor road network model, the starting point and end point of each path are determined according to the location coordinates and geometric shape of each road, and the nodes connected by the starting point and end point of each path are determined according to the 3D building model of each building. Each node and the path between nodes are connected into a network to obtain the outdoor road network model of the smart park. The outdoor road network model of the smart park is scaled at the target scaling ratio, and the 3D building models of each building indicated by each node in the outdoor road network model are integrated with the outdoor road network model to obtain the 3D park model of the smart park.

[0129] In one possible embodiment, see Figure 4 As shown, Figure 4 A flowchart of a method for dividing a region provided by a first embodiment of the present invention is shown, wherein the component information includes a component type, the component type is emergency equipment or non-emergency equipment, and the road type is a safe passage or a non-safe passage; the 3D model of the smart park is divided into risk areas and non-risk areas according to the component information of each building component and the road type of each road, including steps S401 to S403:

[0130] S401: Determine whether the component type of each building component is an emergency device, and determine whether the road type of each road is a safe passage.

[0131] S402: Expand the area with the location of each emergency building component as the center point to obtain a number of risk sub-areas, and expand the area of ​​each emergency road to obtain a number of risk sub-roads, wherein the emergency building component is a building component whose component type is emergency equipment, and the emergency road is a road whose road type is a safe passage.

[0132] S403: Mark each risk sub-area and each risk sub-road in the park 3D model to obtain the risk area, and mark the area other than the risk area in the park 3D model as the non-risk area.

[0133] Specifically, when the area is expanded with the location of each emergency building component as the center point, the scope of the area expansion can be determined according to the properties of the emergency building component. When the emergency level corresponding to the emergency building component is higher, the scope of the area expansion is larger, and the expanded risk sub-areas are larger. When the area is expanded for each emergency road, the scope of the area expansion can be determined according to the properties of the emergency road. When the width of the emergency road (safe passage) is smaller, the scope of the area expansion is larger, and the expanded risk sub-roads are larger.

[0134] In one possible embodiment, see Figure 5 As shown, Figure 5 A flowchart of a method for generating a 3D model of a person provided in Embodiment 1 of the present invention is shown, wherein the method of performing 3D modeling on non-risk persons located in each non-risk area based on each non-risk area image to obtain a 3D model of each non-risk person includes steps S501 to S505:

[0135] S501: For each non-risk area image, a feature extraction model is used to extract static feature data of the non-risk area image to obtain human feature data and background feature data of the non-risk area image.

[0136] Specifically, the human body feature data is used to represent the human body image features in the corresponding image in the form of vector data, and the background feature data is used to represent the background image features in the corresponding image in the form of vector data.

[0137] S502: Generate a human sub-image and a background sub-image of the non-risk area image based on the human feature data and background feature data of the non-risk area image respectively.

[0138] Specifically, each human feature data and each background feature data of the non-risk area image are respectively input into the trained stable diffusion model to obtain a human sub-image of the non-risk area image and a background sub-image.

[0139] S503: Smoothing the image boundary of the human body sub-image of the non-risk area image according to the image boundary pixel value of the background sub-image of the non-risk area image to obtain the target human body sub-image of the non-risk area image.

[0140] Specifically, whether there is a boundary interruption area is determined according to the image boundary pixel value of the background sub-image of the non-risk area image; the image boundary of the human body sub-image adjacent to the boundary interruption area in the background sub-image is gray-filled to obtain the target human body sub-image.

[0141] S504: performing human body key point recognition on each target human body sub-image to obtain human body modeling data of each target human body sub-image.

[0142] Specifically, the human body key point recognition model is used to perform human body key point recognition on the target human body sub-image to obtain the serial number and coordinates of each human body key point in the target human body sub-image, and the serial number and coordinates of each human body key point are determined as human body modeling data.

[0143] S505: Perform 3D modeling based on the human body modeling data of each target human body sub-image to obtain a 3D model of each non-risk person corresponding to each target human body sub-image.

[0144] Specifically, 3D modeling is performed based on the serial number and coordinates of each key point of the human body to obtain a 3D model of each non-risk person.

[0145] In one possible embodiment, see Figure 6 As shown, Figure 6 A flowchart of a prediction 3D model generation method provided by the first embodiment of the present invention is shown, wherein the prediction of the 3D model of each non-risk person at the target time is performed based on the 3D model of each non-risk person at different times to obtain the prediction 3D model of each non-risk person, including steps S601-S603:

[0146] S601: For each risk person, the current action and action change information of the non-risk person is determined based on the 3D model of the non-risk person at different times.

[0147] Specifically, the 3D model of the non-risk personnel at different times is obtained, and the key point serial number and coordinates of each key point of the human body of the non-risk personnel at different times are collected. For each key point of the human body with the same serial number, the displacement distance of the key point of the human body at different times is calculated, and the displacement speed of the key point of the human body is calculated according to the time difference at different times and the displacement distance of the key point of the human body. The displacement information and displacement speed of each key point of the human body are input into the human body motion determination model to determine the current motion of the risk personnel, and the average displacement speed of each key point of the human body is calculated to obtain the motion change information.

[0148] S602: Determine the predicted action of the non-risk person at the target moment according to the current action and action change information of the non-risk person.

[0149] Specifically, the time difference between the target moment and the current moment is calculated, and the displacement distance of each key point of the human body at the target moment is calculated based on the time difference and the coordinates of each key point of the human body at the current moment and the speed of movement change. The displacement distance and displacement speed of each key point of the human body at the target moment are input into the human body movement determination model to determine the predicted movement of the risk person.

[0150] S603: Constructing a predicted 3D model of the non-risk person according to the predicted action of the non-risk person at the target moment.

[0151] Specifically, based on the predicted action of the non-risk person at the target moment, the serial number and coordinates of each key point of the human body at the target moment are determined, and the predicted 3D model of the non-risk person is constructed according to the serial number and coordinates of each key point of the human body at the target moment.

[0152] In one possible embodiment, see Figure 7 As shown, Figure 7A flow chart of an identity authentication method provided in Embodiment 1 of the present invention is shown, wherein the identity information of risk personnel in each risk area is determined based on the risk area image of each risk area, and the identity authentication of the target non-risk personnel is performed according to the identity information of each risk personnel, including steps S701~S702:

[0153] S701: For each risk area, face recognition is performed on the risk area image of the risk area to determine the facial feature value of the recognized face, and the facial feature value of the recognized face is used as the identity information of the risk person corresponding to the face in the risk area.

[0154] Specifically, when performing face recognition on the risk area image of the risk area, the trained face recognition model can be used to obtain the coordinates of several key face points in the risk area image. The face feature value of the corresponding face is calculated based on the coordinates of each key face point as the identity information of the risk person corresponding to the face.

[0155] S702: Obtain facial feature values ​​of the target non-risk person, determine whether the facial feature values ​​of each risk person are the same as the facial feature values ​​of the target non-risk person, and perform identity verification on the target non-risk person based on the determination result.

[0156] Specifically, if the facial feature value of the target non-risk person is the same as at least one of the facial feature values ​​of each risk person, it means that the target non-risk person is one of the risk persons.

[0157] Embodiment 2

[0158] See also Figure 8 As shown, Figure 8 The schematic diagram of the structure of a smart park risk warning system based on 3D modeling provided by the second embodiment of the present invention is shown, wherein the system includes an information collection module 801, a 3D modeling module 802, a risk assessment module 803 and a risk alarm module 804, and the risk alarm module 804 includes a park alarm device 8041 and a remote alarm device 8042;

[0159] The information collection module is used to obtain building information of each building in the smart park and road information of each road in the smart park, wherein the building information includes component information of each building component and a first topological relationship between each building component, and the road information includes a road type of each road and a second topological relationship between each road and each building;

[0160] The 3D modeling module is used to perform 3D modeling on each building based on the component information of each building component in each building and the first topological relationship between each building component to obtain a 3D building model of each building, and perform 3D modeling on the smart park based on the 3D building model of each building, the road type of each road and the second topological relationship between each road and each building to obtain a 3D park model of the smart park;

[0161] The information acquisition module is used to divide the 3D model of the smart park into risk areas and non-risk areas according to the component information of each building component and the road type of each road, and obtain the risk area image of each risk area and the non-risk area image of each non-risk area in real time;

[0162] The 3D modeling module is used to perform 3D modeling on non-risk personnel located in each non-risk area based on each non-risk area image to obtain a 3D model of each non-risk person, and predict the 3D model of each non-risk person at a target time based on the 3D model of each non-risk person at different times to obtain a predicted 3D model of each non-risk person;

[0163] The risk assessment module is used to integrate the predicted 3D models of each non-risk person with the park 3D model, and determine whether the predicted 3D model of each non-risk person appears in the risk area of ​​the park 3D model according to the integration result;

[0164] The risk alarm module is used to issue a park risk warning for the target non-risk person and the risk area through the park alarm device if there is a target non-risk person whose 3D model is predicted to appear in the risk area of ​​the park 3D model;

[0165] The risk assessment module is used to determine the identity information of the risk persons in each risk area based on the risk area images of each risk area at the target time, and to authenticate the target non-risk person according to the identity information of each risk person, and to determine whether the target non-risk person is one of the risk persons according to the authentication result;

[0166] If the target non-risk person is one of the risk persons, the risk alarm module will issue a park risk alarm for the target non-risk person and the risk area through the remote alarm device.

[0167] In a feasible implementation manner, the 3D modeling of each building based on the component information of each building component in each building and the first topological relationship between each building component to obtain the 3D model of each building includes:

[0168] For each building, the coordinates of each building component and the connected space in the building are determined according to the component information of each building component in the building and the first topological relationship between the building components;

[0169] An indoor road network model of the building is constructed according to the coordinate components of each building component and the connected space in the building;

[0170] A 3D building model of the building is constructed according to the indoor road network model of the building, the scaling ratio of the indoor road network model and the types of each building component.

[0171] In a feasible implementation scheme, the 3D modeling of the smart park is performed based on the 3D building model of each building, the road type of each road, and the second topological relationship between each road and each building to obtain the 3D park model of the smart park, including:

[0172] Adjusting the scaling ratio of the 3D building models of the various buildings so that the 3D building models of the various buildings have the same target scaling ratio;

[0173] Generate an outdoor road network model of the smart park according to the road type of each road and the second topological relationship between each road and each building;

[0174] A 3D model of the smart park is constructed based on the 3D building models of each building, the outdoor road network model and the target scaling ratio.

[0175] In a feasible implementation scheme, the component information includes a component type, the component type is emergency equipment or non-emergency equipment, and the road type is a safe passage or a non-safe passage; the 3D model of the smart park is divided into risk areas and non-risk areas according to the component information of each building component and the road type of each road, including:

[0176] Determine whether the component type of each building component is an emergency device, and determine whether the road type of each road is a safe passage;

[0177] Taking the location of each emergency building component as the center point, regional expansion is performed to obtain a number of risk sub-regions, and regional expansion is performed on each emergency road to obtain a number of risk sub-roads, wherein the emergency building component is a building component whose component type is emergency equipment, and the emergency road is a road whose road type is a safe passage;

[0178] Each risk sub-area and each risk sub-road is marked in the park 3D model to obtain the risk area, and the area other than the risk area in the park 3D model is marked as the non-risk area.

[0179] In a feasible implementation manner, the 3D modeling of the non-risk personnel located in each non-risk area based on each non-risk area image to obtain the 3D model of each non-risk personnel includes:

[0180] For each non-risk area image, the static feature data of the non-risk area image is extracted using a feature extraction model to obtain human feature data and background feature data of the non-risk area image;

[0181] Generate a human sub-image and a background sub-image of the non-risk area image based on the human feature data and background feature data of the non-risk area image respectively;

[0182] According to the image boundary pixel value of the background sub-image of the non-risk area image, the image boundary of the human body sub-image of the non-risk area image is smoothed to obtain the target human body sub-image of the non-risk area image;

[0183] Performing human body key point recognition on each target human body sub-image to obtain human body modeling data of each target human body sub-image;

[0184] 3D modeling is performed based on the human body modeling data of each target human body sub-image to obtain a 3D model of each non-risk person corresponding to each target human body sub-image.

[0185] In a feasible implementation scheme, the 3D model of each non-risk person at the target time is predicted based on the 3D model of each non-risk person at different times to obtain the predicted 3D model of each non-risk person, including:

[0186] For each risk person, the current action and action change information of the non-risk person are determined based on the 3D model of the non-risk person at different times;

[0187] Determine the predicted action of the non-risk person at the target moment according to the current action and action change information of the non-risk person;

[0188] A predicted 3D model of the non-risk person is constructed according to the predicted actions of the non-risk person at the target moment.

[0189] In a feasible implementation scheme, the step of determining the identity information of risk personnel in each risk area based on the risk area image of each risk area, and performing identity authentication on the target non-risk personnel according to the identity information of each risk personnel includes:

[0190] For each risk area, face recognition is performed on the risk area image of the risk area to determine the face feature value of the recognized face, and the face feature value of the recognized face is used as the identity information of the risk person corresponding to the face in the risk area;

[0191] Obtain the facial feature value of the target non-risk person, determine whether the facial feature value of each risk person is the same as the facial feature value of the target non-risk person, and authenticate the identity of the target non-risk person based on the determination result.

[0192] Embodiment 3

[0193] Based on the same application concept, see Fig. 9 As shown, Fig. 9 FIG. 4 shows a schematic diagram of the structure of a computer device provided by Embodiment 3 of the present invention, wherein: Fig. 9 As shown, a computer device 900 provided in Embodiment 3 of the present application includes:

[0194] A processor 901, a memory 902 and a bus 903, wherein the memory 902 stores machine-readable instructions executable by the processor 901. When the computer device 900 is running, the processor 901 communicates with the memory 902 via the bus 903. When the processor 901 is running, the machine-readable instructions execute the steps of the smart park risk warning method based on 3D modeling shown in the above-mentioned embodiment 1.

[0195] Embodiment 4

[0196] Based on the same application concept, an embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the smart park risk warning method based on 3D modeling described in any one of the above embodiments are executed.

[0197] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system and device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0198] The computer program product for performing 3D modeling-based smart park risk warning provided in an embodiment of the present invention includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the method described in the previous method embodiment. The specific implementation can be found in the method embodiment, which will not be repeated here.

[0199] The smart park risk warning system based on 3D modeling provided in the embodiment of the present invention can be specific hardware on the device or software or firmware installed on the device. The system provided in the embodiment of the present invention has the same implementation principle and technical effects as the aforementioned method embodiment. For the sake of brief description, for matters not mentioned in the system embodiment, reference can be made to the corresponding contents in the aforementioned method embodiment. Technical personnel in the relevant field can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, devices and units described above can all refer to the corresponding processes in the aforementioned method embodiment, and will not be repeated here.

[0200] In the embodiments provided by the present invention, it should be understood that the disclosed systems and methods can be implemented in other ways. The system embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some communication interface, device or unit, which can be electrical, mechanical or other forms.

[0201] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0202] In addition, each functional unit in the embodiment provided by the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0203] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a service device, or a network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0204] It should be noted that similar numbers and letters represent similar items in the following figures. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. In addition, the terms "first", "second", "third", etc. are only used to distinguish the description and are not to be understood as indicating or implying relative importance.

[0205] Finally, it should be noted that the above-described embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The protection scope of the present invention is not limited thereto. Although the present invention is described in detail with reference to the above-described embodiments, those skilled in the art should understand that any person skilled in the art can still modify the technical solutions described in the above-described embodiments within the technical scope disclosed by the present invention, or can easily think of changes, or perform equivalent replacements on some of the technical features thereof; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention. They should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A smart park risk warning method based on 3D modeling, characterized in that: Applied to a smart park risk early warning system based on 3D modeling, the system includes an information collection module, a 3D modeling module, a risk assessment module and a risk alarm module, the risk alarm module includes a park alarm device and a remote alarm device, and the method includes: The information collection module obtains the building information of each building in the smart park and the road information of each road in the smart park, wherein the building information includes the component information of each building component and the first topological relationship between each building component, and the road information includes the road type of each road and the second topological relationship between each road and each building; The 3D modeling module performs 3D modeling on each building based on the component information of each building component in each building and the first topological relationship between each building component to obtain a 3D building model of each building, and performs 3D modeling on the smart park based on the 3D building model of each building, the road type of each road and the second topological relationship between each road and each building to obtain a 3D park model of the smart park; The information acquisition module divides the 3D model of the smart park into risk areas and non-risk areas according to the component information of each building component and the road type of each road, and obtains the risk area image of each risk area and the non-risk area image of each non-risk area in real time; The 3D modeling module performs 3D modeling on non-risk personnel located in each non-risk area based on each non-risk area image to obtain a 3D model of each non-risk person, and predicts the 3D model of each non-risk person at a target time based on the 3D model of each non-risk person at different times to obtain a predicted 3D model of each non-risk person; The risk assessment module integrates the predicted 3D models of each non-risk person with the park 3D model, and determines whether the predicted 3D models of each non-risk person appear in the risk area of ​​the park 3D model according to the integration result; If there is a target non-risk person whose 3D model is predicted to appear in the risk area of ​​the park 3D model, the risk alarm module issues a park risk warning for the target non-risk person and the risk area through the park alarm device; At the target moment, the risk assessment module determines the identity information of the risk persons in each risk area based on the risk area images of each risk area, and performs identity authentication on the target non-risk person according to the identity information of each risk person, and determines whether the target non-risk person is one of the risk persons according to the identity authentication result; If the target non-risk person is one of the risk persons, the risk alarm module issues a park risk alarm for the target non-risk person and the risk area through the remote alarm device; The method of performing 3D modeling on each building based on the component information of each building component in each building and the first topological relationship between each building component to obtain a 3D building model of each building includes: For each building, the coordinates of each building component and the connected space in the building are determined according to the component information of each building component in the building and the first topological relationship between the building components; An indoor road network model of the building is constructed according to the coordinate components of each building component and the connected space in the building; A 3D building model of the building is constructed according to the indoor road network model of the building, the scaling ratio of the indoor road network model and the types of each building component.

2. The method according to claim 1, characterized in that The 3D modeling of the smart park is performed based on the 3D building models of each building, the road type of each road, and the second topological relationship between each road and each building to obtain the 3D park model of the smart park, including: Adjusting the scaling ratio of the 3D building models of the various buildings so that the 3D building models of the various buildings have the same target scaling ratio; Generate an outdoor road network model of the smart park according to the road type of each road and the second topological relationship between each road and each building; A 3D model of the smart park is constructed based on the 3D building models of each building, the outdoor road network model and the target scaling ratio.

3. The method according to claim 1, characterized in that The component information includes a component type, the component type is emergency equipment or non-emergency equipment, and the road type is a safe passage or a non-safe passage; the 3D model of the smart park is divided into risk areas and non-risk areas according to the component information of each building component and the road type of each road, including: Determine whether the component type of each building component is an emergency device, and determine whether the road type of each road is a safe passage; Taking the location of each emergency building component as the center point, regional expansion is performed to obtain a number of risk sub-regions, and regional expansion is performed on each emergency road to obtain a number of risk sub-roads, wherein the emergency building component is a building component whose component type is emergency equipment, and the emergency road is a road whose road type is a safe passage; Each risk sub-area and each risk sub-road is marked in the park 3D model to obtain the risk area, and the area other than the risk area in the park 3D model is marked as the non-risk area.

4. The method according to claim 1, characterized in that: The 3D modeling of the non-risk personnel located in each non-risk area based on each non-risk area image to obtain the 3D model of each non-risk personnel includes: For each non-risk area image, the static feature data of the non-risk area image is extracted using a feature extraction model to obtain human feature data and background feature data of the non-risk area image; Generate a human sub-image and a background sub-image of the non-risk area image based on the human feature data and background feature data of the non-risk area image respectively; According to the image boundary pixel value of the background sub-image of the non-risk area image, the image boundary of the human body sub-image of the non-risk area image is smoothed to obtain the target human body sub-image of the non-risk area image; Performing human body key point recognition on each target human body sub-image to obtain human body modeling data of each target human body sub-image; 3D modeling is performed based on the human body modeling data of each target human body sub-image to obtain a 3D model of each non-risk person corresponding to each target human body sub-image.

5. The method according to claim 1, characterized in that The 3D model of each non-risk person at the target time is predicted based on the 3D model of each non-risk person at different times to obtain the predicted 3D model of each non-risk person, including: For each risk person, the current action and action change information of the non-risk person are determined based on the 3D model of the non-risk person at different times; Determine the predicted action of the non-risk person at the target moment according to the current action and action change information of the non-risk person; A predicted 3D model of the non-risk person is constructed according to the predicted actions of the non-risk person at the target moment.

6. The method according to claim 1, characterized in that The step of determining the identity information of the risk personnel in each risk area based on the risk area image of each risk area, and performing identity authentication on the target non-risk personnel according to the identity information of each risk personnel includes: For each risk area, face recognition is performed on the risk area image of the risk area to determine the face feature value of the recognized face, and the face feature value of the recognized face is used as the identity information of the risk person corresponding to the face in the risk area; Obtain the facial feature value of the target non-risk person, determine whether the facial feature value of each risk person is the same as the facial feature value of the target non-risk person, and authenticate the identity of the target non-risk person based on the determination result.

7. A smart park risk warning system based on 3D modeling, characterized in that: The system includes an information collection module, a 3D modeling module, a risk assessment module and a risk alarm module, and the risk alarm module includes a park alarm device and a remote alarm device; The information collection module is used to obtain building information of each building in the smart park and road information of each road in the smart park, wherein the building information includes component information of each building component and a first topological relationship between each building component, and the road information includes a road type of each road and a second topological relationship between each road and each building; The 3D modeling module is used to perform 3D modeling on each building based on the component information of each building component in each building and the first topological relationship between each building component to obtain a 3D building model of each building, and perform 3D modeling on the smart park based on the 3D building model of each building, the road type of each road and the second topological relationship between each road and each building to obtain a 3D park model of the smart park; The information acquisition module is used to divide the 3D model of the smart park into risk areas and non-risk areas according to the component information of each building component and the road type of each road, and obtain the risk area image of each risk area and the non-risk area image of each non-risk area in real time; The 3D modeling module is used to perform 3D modeling on non-risk personnel located in each non-risk area based on each non-risk area image to obtain a 3D model of each non-risk person, and predict the 3D model of each non-risk person at a target time based on the 3D model of each non-risk person at different times to obtain a predicted 3D model of each non-risk person; The risk assessment module is used to integrate the predicted 3D models of each non-risk person with the park 3D model, and determine whether the predicted 3D model of each non-risk person appears in the risk area of ​​the park 3D model according to the integration result; The risk alarm module is used to issue a park risk warning for the target non-risk person and the risk area through the park alarm device if there is a target non-risk person whose 3D model is predicted to appear in the risk area of ​​the park 3D model; The risk assessment module is used to determine the identity information of the risk persons in each risk area based on the risk area images of each risk area at the target time, and to authenticate the target non-risk person according to the identity information of each risk person, and to determine whether the target non-risk person is one of the risk persons according to the authentication result; If the target non-risk person is one of the risk persons, the risk alarm module issues a park risk alarm for the target non-risk person and the risk area through the remote alarm device; The method of performing 3D modeling on each building based on the component information of each building component in each building and the first topological relationship between each building component to obtain a 3D building model of each building includes: For each building, the coordinates of each building component and the connected space in the building are determined according to the component information of each building component in the building and the first topological relationship between the building components; An indoor road network model of the building is constructed according to the coordinate components of each building component and the connected space in the building; A 3D building model of the building is constructed according to the indoor road network model of the building, the scaling ratio of the indoor road network model and the types of each building component.

8. A computer device, characterized in that: include: A processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor. When the computer device is running, the processor and the memory communicate via the bus. When the machine-readable instructions are executed by the processor, the steps of the smart park risk warning method based on 3D modeling are performed.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which, when executed by a processor, executes the steps of the 3D modeling-based smart park risk warning method as described in any one of claims 1 to 6.

Images