A method and system for intelligent detection and early warning of falling objects from high altitude buildings

Abstract

This application relates to a method and system for intelligent detection and early warning of falling objects from buildings, and relates to the technical field of smart cities. The method comprises obtaining a picture captured by a detection unit; determining whether a falling object exists based on the picture; if so, performing image analysis on the picture to obtain an analysis result; and controlling an alarm unit to issue an early warning based on the analysis result. The method has the effect of timely detecting and early warning of falling objects from high altitudes, thereby reducing the possibility of injury to the human body caused by falling objects from high altitudes.

Description

Technical Field

[0001] The present application relates to the technical field of smart cities, and in particular to a method and system for intelligent detection and early warning of falling objects from high altitudes in buildings. Background Art

[0002] Throwing objects from high places is called "the pain hanging over the city." This behavior is not only uncivilized, but also poses a huge hidden danger to social security. Moreover, since most of the places where the incidents occur are high-rise, there are few witnesses and the throwing time is short, it is difficult for law enforcement agencies to hold the throwers accountable after the incident. As a result, such throwing incidents are common and cannot be stopped.

[0003] Objects thrown from high places are a common problem in residential communities, and their management and rectification are crucial to public security and the safety of residents. However, relying solely on traditional ideological and moral education cannot achieve very effective results, especially in high-rise residential buildings, where people throw objects for convenience, which is even more dangerous. Scientific calculations and testing have shown that a 30-gram egg thrown from a height of approximately 12 meters can cause boils, while a drop of approximately 54 meters can fracture a person's skull. A drop of 75 meters can cause instant death, causing irreparable damage.

[0004] In related technologies, cameras are generally used for monitoring, but cameras can usually only record videos and cannot actively identify and warn. Manual video review is still required for identification, and timely detection and warning cannot be achieved. Therefore, there is an urgent need for an intelligent detection and warning method for falling objects from high altitudes in buildings. Summary of the Invention

[0005] In order to detect and warn of falling objects from high altitudes in a timely manner, the present application provides an intelligent detection and warning method and system for falling objects from high altitudes in buildings.

[0006] In the first aspect, the present application provides an intelligent detection and early warning method for falling objects from high altitude buildings, which adopts the following technical solutions:

[0007] An intelligent detection and early warning method for falling objects from high altitude buildings, comprising:

[0008] Obtaining the images captured by the detection unit;

[0009] Determining whether there is a falling object from a high altitude based on the captured image;

[0010] If yes, perform image analysis on the captured image to obtain analysis results;

[0011] Based on the analysis result, the alarm unit is controlled to issue an early warning.

[0012] By adopting the above technical solution, when there are falling objects in the shooting picture, it is necessary to analyze the falling objects. If the falling objects have not fallen, an early warning can be issued. If the falling objects have fallen, an alarm can be sounded in time to remind passers-by, thereby reducing the possibility of human injury caused by falling objects.

[0013] Optionally, performing image analysis on the captured image to obtain an analysis result includes:

[0014] Obtaining a background image captured by the detection unit;

[0015] Performing image comparison between the background image and the captured image;

[0016] Determining the position of the falling object within the captured image;

[0017] Amplifying the captured image based on the position of the falling object to obtain an amplified image;

[0018] determining the type of the falling object based on the magnified image;

[0019] Determining whether the falling object causes harm to the human body based on the type of the falling object;

[0020] If so, the type of falling objects and the judgment result are used as analysis results, and the judgment result includes whether the falling objects cause harm to the human body.

[0021] By adopting the above technical solution, the background picture and the shooting picture are compared to know whether there are falling objects in the shooting picture. If there are falling objects, the position of the falling objects in the shooting picture can be magnified to more clearly identify the type of falling objects, so as to conduct a more detailed analysis of the falling objects.

[0022] Optionally, after determining the type of the falling object based on the magnified image, the method further includes:

[0023] Obtaining support information of the high-altitude falling object;

[0024] Determining the stability of the falling object based on the support information;

[0025] Obtaining weather information of the environment where the high-altitude falling object is located;

[0026] Determining a first probability of the falling object occurring based on the stability and the weather information;

[0027] The first probability of falling is used as an analysis result.

[0028] By adopting the above technical solution, the stability of falling objects from high altitude can be analyzed by obtaining the support information of falling objects from high altitude, so as to calculate the first probability of falling objects from high altitude. According to the first probability, early warning can be achieved, thereby reducing the possibility of falling objects from high altitude causing harm to the human body.

[0029] Optionally, performing image analysis on the captured image to obtain an analysis result further includes:

[0030] Acquire a historical shooting picture of the current detection unit based on the shooting picture;

[0031] Counting the number of high-altitude falling objects based on the historical footage;

[0032] Calculating a second probability of a high-altitude falling object phenomenon occurring in the area where the current detection unit is located based on the number of occurrences of the high-altitude falling object phenomenon;

[0033] determining a danger level of the area where the detection unit is currently located based on the first probability and the second probability;

[0034] The risk level is used as the analysis result.

[0035] By adopting the above technical solution and counting the number of falling objects from high altitude in the history of buildings, the probability of falling objects from high altitude can be further analyzed, making the predicted probability of falling objects from high altitude more accurate, thereby providing more accurate early warning and reducing the risk of falling objects from high altitude.

[0036] Optionally, controlling the alarm unit to issue an early warning based on the analysis result includes:

[0037] determining accident-frequently occurring areas of the building based on the analysis results;

[0038] Determining an alarm type based on the accident-frequently occurring area;

[0039] determining an alarm mode based on the alarm type;

[0040] An early warning is issued based on the alarm method.

[0041] Optionally, the alarm type includes an emergency alarm, and determining the alarm mode based on the alarm type includes:

[0042] When the alarm type is an emergency alarm, obtaining the initial position of the high-altitude falling object;

[0043] Calculating the falling time of the high-altitude falling object based on the initial position;

[0044] Determining whether the falling time is greater than a preset time;

[0045] If the falling time is greater than the preset time, controlling the interception unit to intercept the falling object;

[0046] If the falling time is not greater than the preset time, the alarm unit is controlled to issue an alarm.

[0047] By adopting the above technical solution, a more effective alarm method can be selected according to the falling time of the falling object, which not only saves resources, but also effectively warns passers-by and reduces the possibility of passers-by being injured.

[0048] Optionally, the controlling and intercepting unit intercepting the falling object includes:

[0049] constructing a digital twin model based on the building;

[0050] Predicting the falling trajectory of the high-altitude falling object based on the digital twin model;

[0051] Determine an intersection point of the falling object within a preset height based on the falling trajectory;

[0052] Obtaining the position information and movement speed of all the interception units;

[0053] determining a moving distance of the intercepting unit based on the position information of the intercepting unit and the intersection point;

[0054] determining a moving time of each of the intercepting units based on the moving distance and the moving speed;

[0055] The interception unit is controlled based on the movement time to intercept the falling object.

[0056] By adopting the above technical solution, the moving distance of the interception unit can be calculated according to the position of the intersection and the position information of the interception unit, and then the moving time can be determined according to the moving distance and the moving speed, so as to judge in advance whether the interception unit can intercept falling objects from high altitude. If it cannot be intercepted, other alarm methods are used to remind passers-by, thereby reducing the possibility of falling objects from high altitude causing damage to the human body.

[0057] Optionally, before controlling the alarm unit to issue an early warning based on the analysis result, the method further includes:

[0058] Obtaining the flow of people in the building during each preset time period;

[0059] Sorting the flow of people in each preset time period to obtain a sorting result;

[0060] Formulate a manual inspection strategy based on the ranking results;

[0061] Determine the location of each inspection terminal based on the manual inspection strategy;

[0062] Determine the location of the falling object based on the analysis result;

[0063] Formulate an inspection route based on the location of the falling object and the location of each inspection terminal;

[0064] The inspection route is sent to the inspection terminal.

[0065] In the second aspect, the present application provides an intelligent detection and warning system for falling objects from high altitude buildings, which adopts the following technical solutions:

[0066] An intelligent detection and warning system for falling objects from a high altitude in a building, applied to the intelligent detection and warning method for falling objects from a high altitude in a building according to any one of the first aspects, comprises a detection unit, a control unit, and an alarm unit; the detection unit and the alarm unit are both electrically connected to the control unit;

[0067] The detection unit is used to capture images of the building;

[0068] The control unit is used to obtain the shooting picture taken by the detection unit; determine whether there is a falling object from a high altitude based on the shooting picture; if so, perform image analysis on the shooting picture to obtain an analysis result; and control the alarm unit to issue an early warning based on the analysis result;

[0069] The alarm unit is used to issue an early warning.

[0070] By adopting the above technical solution, when there are falling objects in the shooting picture, it is necessary to analyze the falling objects. If the falling objects have not fallen, an early warning can be issued. If the falling objects have fallen, an alarm can be sounded in time to remind passers-by, thereby reducing the possibility of human injury caused by falling objects.

[0071] Optionally, the system further includes a communication unit and an interception unit; the control unit is wirelessly connected to the interception unit via the communication unit; and further includes:

[0072] The inspection terminal wirelessly connected to the control unit is used to receive the inspection route.

[0073] In summary, this application includes at least one of the following beneficial technical effects:

[0074] 1. When there are falling objects in the captured image, it is necessary to analyze the falling objects. If the falling objects have not fallen, an early warning can be issued. If the falling objects have fallen, an alarm can be issued in time to alert passers-by, thereby reducing the possibility of injury from falling objects.

[0075] 2. By adopting the above technical solution, the moving distance of the interception unit can be calculated according to the location of the intersection and the position information of the interception unit, and then the moving time can be determined according to the moving distance and moving speed, so as to judge in advance whether the interception unit can intercept falling objects from high altitude. If it cannot be intercepted, other alarm methods are used to remind passers-by, thereby reducing the possibility of injury to the human body caused by falling objects from high altitude. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] Figure 1 This is a structural block diagram of an intelligent detection and early warning system for falling objects from high altitudes in buildings provided in an embodiment of the present application.

[0077] Figure 2 This is a flow chart of an intelligent detection and early warning method for falling objects from high altitudes in buildings provided in an embodiment of the present application.

[0078] Explanation of the accompanying symbols: 1. Detection unit; 2. Control unit; 3. Alarm unit; 4. Communication unit; 5. Interception unit; 6. Inspection terminal. DETAILED DESCRIPTION

[0079] In order to make the purpose, technical solutions and advantages of this application more clear, the following Figure 1-2 It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application.

[0080] The embodiment of the present application discloses an intelligent detection and warning system for falling objects from high altitude buildings. Figure 1 The intelligent detection and warning system for falling objects from high altitude buildings includes a detection unit 1, a control unit 2 and an alarm unit 3. The detection unit 1 and the alarm unit 3 are both electrically connected to the control unit 2.

[0081] In this embodiment, the detection unit 1 includes a plurality of network cameras placed on the ground and shooting at an elevation angle, which are used to monitor and shoot the building from multiple angles, and transmit the obtained shooting images and monitoring images to the control unit 2 in real time.

[0082] As an optional implementation of this embodiment, the control unit 2 can be a server or a terminal device, wherein the server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. The terminal device can be a smartphone, a tablet computer, a desktop computer, etc., but is not limited thereto.

[0083] The alarm unit 3 includes a buzzer alarm and an audio player, wherein the buzzer alarm and the audio player are both electrically connected to the control unit 2 .

[0084] When the detection unit 1 detects that there is a falling object on the building, the control unit 2 controls the buzzer alarm and the audio player to start, thereby reminding people and reducing the possibility of causing damage to people.

[0085] In this embodiment, the intelligent detection system for falling objects from high altitudes in buildings also includes a communication unit 4 and an interception unit 5. The interception unit 5 is electrically connected to the control unit 2 through the communication unit 4. When it is necessary to intercept falling objects from high altitudes, the control unit 2 communicates with the interception unit 5 through the communication unit 4 and sends an interception instruction to the interception unit 5. The interception unit 5 intercepts the falling objects from high altitudes according to the received interception instruction. The interception instruction includes the falling trajectory and falling time of the falling objects from high altitudes.

[0086] As an optional implementation of this embodiment, the communication unit 4 can be a device that communicates through the network, or a device that communicates through a radio station. The interception unit 5 can use drone equipment for interception, or install an interception net at a preset height of the building for interception. In this embodiment, drone equipment is used for interception.

[0087] As an optional implementation of this embodiment, the intelligent detection and warning system for falling objects from high altitudes in buildings also includes a patrol terminal 6, which is wirelessly connected to the control unit 2. When manual detection of falling objects from high altitudes in buildings is performed, the staff holds the patrol terminal 6 to observe the situation at high altitudes in real time, and conducts patrols according to the patrol routes prompted by the patrol terminal 6, thereby realizing on-site patrols by the staff and reducing the possibility of accidents caused by untimely patrols.

[0088] Preferably, the inspection terminal 6 can be a smart phone, iPad, tablet computer or computer, but is not limited thereto.

[0089] Reference Figure 2 The embodiment of the present application also provides a method for intelligent detection and early warning of falling objects from high altitudes in buildings. The main process of the method is described as follows (steps S101-S104):

[0090] Step S101: Acquire a picture captured by the detection unit 1.

[0091] Specifically, due to the randomness and low probability of falling objects from high altitudes, it is necessary to photograph and monitor buildings in real time. Therefore, buildings can be monitored in real time through network cameras. The control unit 2 extracts the shooting images to obtain shooting images of the building so as to detect abnormal conditions of the building at any time, that is, whether there are falling objects from high altitudes in the building.

[0092] Step S102, based on the captured image, determine whether there is any falling object from a high altitude. If so, proceed to step S103; if not, repeat step S101.

[0093] In this embodiment, the shooting picture is input into the convolutional neural network model, and the feature extraction of the shooting picture is performed by the convolutional neural network model. Then, it is judged whether there is a falling object from a high altitude based on the extracted features. For example, the feature information of a banana peel is extracted from the shooting picture. At this time, it can be judged from the shooting picture that a person is holding a banana peel and performing a throwing action, and it can be determined that there is a falling object from a high altitude.

[0094] In this embodiment, if there are items stored on the balcony, such as flower pots, in the captured image, it can be determined that there are objects falling from a high altitude.

[0095] Step S103: performing image analysis on the captured image to obtain analysis results.

[0096] Specifically, performing image analysis on the captured image to obtain the analysis results includes: obtaining the background image captured by the detection unit 1; performing image comparison between the background image and the captured image; determining the position of the falling object in the captured image; amplifying the captured image based on the position of the falling object to obtain an amplified image; determining the type of the falling object based on the amplified image; judging whether the falling object causes harm to the human body based on the type of the falling object; if so, taking the type of the falling object and the judgment result as the analysis result, and the judgment result includes whether the falling object causes harm to the human body.

[0097] In this embodiment, a shot picture without the phenomenon of objects being thrown from high altitude is used as the background picture, that is, a shot picture with only buildings is used as the background picture, and then the shot picture and the background picture are compared. When there are inconsistencies between the shot picture and the background picture, the inconsistent places are marked on the shot picture. At the same time, in order to accurately identify the type of falling objects from high altitude, it is necessary to zoom in on the marked position in the shot picture to obtain an enlarged image of the shot picture, and then determine the type of falling objects from high altitude based on the enlarged image. Among them, the types of falling objects from high altitude include heavy objects, garbage, jewelry and other types. For example, heavy objects falling from high altitude include but are not limited to green plants and electrical equipment, garbage objects falling from high altitude include but are not limited to banana peels, apple cores and plastic bottles, jewelry objects falling from high altitude include but are not limited to necklaces, rings and glasses, and other types of falling objects falling from high altitude include but are not limited to plastic bags, mobile phones and human bodies.

[0098] After obtaining the magnified image, it is necessary to obtain a feature image of the marked position based on the magnified image, and then compare the feature image with the feature image in the preset database to determine the type of falling objects. Since heavy falling objects, garbage falling objects and other types of falling objects can cause harm to the human body, when the identified falling object type is any of heavy falling objects, garbage falling objects or other types of falling objects, it is determined that the falling object can cause harm to the human body. At this time, the analysis result includes the type of falling object and the judgment result.

[0099] Furthermore, after determining the type of the falling object based on the magnified image, the method also includes: obtaining support information of the falling object; determining the stability of the falling object based on the support information; obtaining weather information of the environment in which the falling object is located; determining a first probability of the falling object falling based on the stability and weather information; and using the first probability of the falling object as an analysis result.

[0100] In this embodiment, after determining the type of falling objects, if the falling objects are objects placed on the balcony, such as flower pots, it is necessary to perform a support analysis on the balcony, that is, to find out whether the balcony has cracks or damages, and whether there is a support structure to reinforce the balcony by shooting the picture. Whether the balcony has cracks or damages, and whether there is a support structure to reinforce the balcony are used as support information. If there are cracks or damages in the balcony, the flower pots placed on the balcony may be unstable. When in a storm, the possibility of the flower pots falling may increase. Therefore, it is necessary to calculate the first probability of the flower pots falling according to the current weather conditions and the stability of the flower pots, and use the first probability as the analysis result. When the first probability is greater than the preset threshold, it is necessary to dispatch staff to patrol to reduce the risk of the flower pots falling.

[0101] Furthermore, performing image analysis on the captured image to obtain the analysis result also includes: obtaining the historical captured image of the current detection unit 1 based on the captured image; counting the number of times that falling objects occurred based on the historical captured image; calculating the second probability of falling objects occurring in the area where the current detection unit 1 is located based on the number of times that falling objects occurred; determining the danger level of the area where the current detection unit 1 is located based on the first probability and the second probability; and using the danger level as the analysis result.

[0102] In this embodiment, for buildings, although falling objects from high altitudes are random, if there is a phenomenon of man-made falling objects from high altitudes in a building, it is necessary to obtain historical shooting images, count the number of falling objects from high altitudes at the current detection unit 1 based on the historical shooting images, calculate the frequency of falling objects from high altitudes based on the number of falling objects from high altitudes and the time when the falling objects occur, and estimate the probability of falling objects from high altitudes based on the frequency of falling objects from high altitudes.

[0103] In this embodiment, the danger level includes a first danger level, a second danger level and a third danger level. When either the first probability or the second probability is greater than the first preset standard probability, the danger level is the first danger level. When either the first probability or the second probability is between the first preset standard probability and the second preset standard probability, the danger level is the second danger level. When both the first probability and the second probability are less than the second preset standard probability, the danger level is the third danger level.

[0104] For example, if the number of falling objects in the current building is 12 times within a year, then the frequency of falling objects in the current building is 12 times / year. At this time, it can be estimated that the second probability of falling objects occurring each month is 1 / 12, of which the number of falling objects caused by weather reasons is 2 times, that is, the first probability is 1 / 72 at this time. If one percent is the first standard probability, then the first probability and the second probability at this time are both greater than the standard probability, so the danger level of the area where the current detection unit 1 is located is the first danger level; the above is only for evidence and easy understanding.

[0105] Step S104: Control the alarm unit 3 to issue an early warning based on the analysis result.

[0106] Specifically, controlling the alarm unit 3 to issue an early warning based on the analysis result includes: determining accident-prone areas of the building based on the analysis result; determining an alarm type based on the accident-prone areas; determining an alarm method based on the alarm type; and issuing an early warning based on the alarm method.

[0107] In this embodiment, after obtaining the analysis results, since buildings have different floor heights, different alarm methods are used for falling objects at different floor heights, so that effective methods are used to alarm for different falling objects at high altitudes, thereby maximizing resource utilization.

[0108] Furthermore, the alarm type includes an emergency alarm, and determining the alarm method based on the alarm type includes: when the alarm type is an emergency alarm, obtaining the initial position of the falling object from high altitude; calculating the falling time of the falling object from high altitude based on the initial position; judging whether the falling time is greater than the preset time; if the falling time is greater than the preset time, controlling the interception unit 5 to intercept the falling object from high altitude; if the falling time is not greater than the preset time, controlling the alarm unit 3 to issue an alarm.

[0109] In this embodiment, when the type of falling object is a heavy object and the danger level of the falling object is the first danger level, the alarm type at this time is an emergency alarm. The initial position of the falling object is determined by shooting the picture, that is, which floor the falling object is on, and then the falling time of the falling object is calculated according to the initial position, and the alarm method is selected according to the falling time.

[0110] For example, the highest floor of a building is the 15th floor, and loudspeakers are used to warn people from the 3rd to the 5th floor, while interception unit 5 is used to intercept falling objects from the 6th floor and above. When the falling object is a flower pot and is located on the 10th floor, the alarm type at this time is an emergency alarm, and the falling time of the flower pot is obtained by calculation. For example, if the falling time of the flower pot is five seconds, the preset time is the longest time for the interception unit 5 to reach any position in the building at the slowest. In this embodiment, the preset time is 3 seconds. At this time, the falling object can be quickly intercepted by controlling the interception unit 5, thereby reducing the risk of injury to the human body caused by the falling object. When the falling time of the flower pot is 2 seconds, the falling time is less than the preset time. At this time, the interception unit 5 cannot intercept the falling object from the high altitude, so it is necessary to control the alarm unit 3 to sound an alarm to remind nearby people.

[0111] It is worth noting that, in this embodiment, the alarm type also includes normal alarm and delayed alarm. When the type of falling objects from high altitude is any one of garbage-type falling objects, jewelry-type falling objects and other types of falling objects, and the danger level is the second danger level, the alarm type is normal alarm. When the falling objects from high altitude are any one of garbage-type falling objects, jewelry-type falling objects and other types of falling objects, and the danger level is the third danger level, the alarm type is delayed alarm.

[0112] Furthermore, controlling the interception unit 5 to intercept falling objects from high altitude includes: building a digital twin model based on the building; predicting the falling trajectory of falling objects from high altitude based on the digital twin model; determining the intersection of falling objects from high altitude within a preset height based on the falling trajectory; obtaining the position information and moving speed of all interception units 5; determining the moving distance of the interception unit 5 based on the position information and intersection of the interception unit 5; determining the moving time of each interception unit 5 based on the moving distance and moving speed; and controlling the interception unit 5 to intercept falling objects from high altitude based on the moving time.

[0113] In this embodiment, when it is necessary to use the interception unit 5 to intercept falling objects from high altitude, the falling trajectory of the falling objects from high altitude is simulated in the digital twin model, so that the intersection of the falling objects from high altitude and the preset height is determined in the digital twin model. This intersection is the position where the interception unit 5 intercepts the falling objects from high altitude. The moving time of the interception unit 5 is calculated based on the current position information and moving speed of the interception unit 5 and the position of the intersection. Since there are multiple interception units 5 near the building, it is necessary to calculate the moving time of each interception unit 5, sort the moving times, find the interception unit 5 with the shortest moving time, and then control the interception unit 5 with the shortest time to reach the intersection, thereby achieving the interception of falling objects from high altitude.

[0114] For example, in this embodiment, there are three interception units 5. At this time, the intersection is on the third floor, and the location of the high-altitude throwing objects is on the tenth floor. The interception units 5 are respectively on the fifth floor, the ground and the seventh floor. The moving speed of the interception units 5 is 20 meters / second. At this time, the interception unit 5 with the shortest moving distance is the interception unit 5 on the fifth floor, and the moving time is the shortest. Therefore, by controlling the interception unit 5 on the fifth floor to intercept high-altitude falling objects, the possibility of high-altitude falling objects causing harm to the human body can be reduced. At the same time, in this embodiment, although the interception unit 5 on the fifth floor is controlled to intercept, the control unit 2 also needs to control the other two interception units 5 to intercept at the same time, so as to increase the possibility of interception and further reduce the possibility of causing harm to the human body.

[0115] Furthermore, before controlling the alarm unit 3 to issue an early warning based on the analysis results, the method also includes: obtaining the flow of people in each preset time period of the building; sorting the flow of people in each preset time period to obtain a sorting result; formulating a manual inspection strategy based on the sorting result; determining the position of each inspection terminal 6 based on the manual inspection strategy; determining the position of falling objects from high altitude based on the analysis results; formulating the shortest inspection route based on the position of the falling objects from high altitude and the position of each inspection terminal 6; and sending the inspection route to the inspection terminal 6.

[0116] In this embodiment, if the building is a residence in a community, the daily flow of people is basically fixed. For example, the flow of people is heaviest from 7 to 9 in the morning, followed by 7 to 9 in the evening, and the flow of people in other time periods is less than these two time periods. The specific information can be obtained through monitoring video. It is a conventional technical means and will not be elaborated here. Each preset time period is sorted according to the flow of people to obtain the sorting results, and then a manual inspection strategy is formulated and sent to the inspection terminal 6. At this time, the location of each inspection terminal 6 can be obtained through the GPS system. When falling objects occur, the shortest inspection route can be specified and the inspection personnel can be reminded based on the location of the inspection terminal 6 and the location of the falling objects, so that the inspection personnel can quickly reach the location of the falling objects, intercept the falling objects or remind passers-by.

[0117] Since the inspection personnel conduct inspections according to the manual inspection strategy and formulate the shortest inspection strategy according to the location of the inspection terminal 6 and the location of the falling objects, each inspection personnel can quickly reach any location of the building to intercept the falling objects or remind passers-by.

[0118] The above are all preferred embodiments of the present application and are not intended to limit the scope of protection of this application. Unless otherwise specified, any feature disclosed in this specification (including the abstract and drawings) may be replaced by other equivalent or similar features. In other words, unless otherwise specified, each feature is merely an example of a series of equivalent or similar features.

Claims

1. An intelligent detection method for falling objects from high altitude buildings, characterized in that: include: Obtaining a picture taken by the detection unit (1); Determining whether there is a falling object from a high altitude based on the captured image; If yes, perform image analysis on the captured image to obtain analysis results; Controlling the alarm unit to issue an early warning based on the analysis result; The performing image analysis on the captured image to obtain the analysis result includes: Obtaining the background image captured by the detection unit (1); Performing image comparison between the background image and the captured image; Determining the position of the falling object within the captured image; Amplifying the captured image based on the position of the falling object to obtain an amplified image; determining the type of the falling object based on the magnified image; Determining whether the falling object causes harm to the human body based on the type of the falling object; If yes, the type of the falling object and the judgment result are used as the analysis result, wherein the judgment result includes whether the falling object causes harm to the human body; After determining the type of the falling object based on the magnified image, the method further includes: Obtaining support information of the falling object, wherein the support information includes cracks, damage, and support structure of the support; Determining the stability of the falling object based on the support information; Obtaining weather information of the environment where the high-altitude falling object is located; Determining a first probability of the falling object from high altitude based on the stability and the weather information; Taking the first probability of falling as an analysis result; The performing image analysis on the captured image to obtain an analysis result further comprises: Acquire a historical shooting picture of the current detection unit based on the shooting picture; Counting the number of high-altitude falling objects based on the historical footage; Calculating a second probability of a high-altitude falling object phenomenon occurring in the area where the current detection unit is located based on the number of occurrences of the high-altitude falling object phenomenon; determining a danger level of the area where the detection unit is currently located based on the first probability and the second probability; The risk level is used as the analysis result.

2. The method according to claim 1, characterized in that The controlling the alarm unit to issue an early warning based on the analysis result includes: determining accident-frequently occurring areas of the building based on the analysis results; Determining an alarm type based on the accident-frequently occurring area; determining an alarm mode based on the alarm type; An early warning is issued based on the alarm method.

3. The method according to claim 2, characterized in that The alarm type includes an emergency alarm, and determining the alarm mode based on the alarm type includes: When the alarm type is an emergency alarm, obtaining the initial position of the high-altitude falling object; Calculating the falling time of the high-altitude falling object based on the initial position; Determining whether the falling time is greater than a preset time; If the falling time is greater than the preset time, controlling the interception unit (5) to intercept the high-altitude falling object; If the falling time is not greater than the preset time, the alarm unit (3) is controlled to issue an alarm.

4. The method according to claim 3, characterized in that The controlling and intercepting unit intercepting the falling object comprises: constructing a digital twin model based on the building; Predicting the falling trajectory of the high-altitude falling object based on the digital twin model; Determine an intersection point of the falling object within a preset height based on the falling trajectory; Obtaining the position information and movement speed of all the interception units (5); determining a moving distance of the intercepting unit (5) based on the position information of the intercepting unit (5) and the intersection; Determining the moving time of each of the intercepting units (5) based on the moving distance and the moving speed; The interception unit (5) is controlled based on the movement time to intercept the high-altitude falling object.

5. The method according to claim 1, wherein Before controlling the alarm unit to issue an early warning based on the analysis result, the method further includes: Obtaining the flow of people in the building during each preset time period; Sorting the flow of people in each preset time period to obtain a sorting result; Formulate a manual inspection strategy based on the ranking results; Determining the location of each inspection terminal (6) based on the manual inspection strategy; Determine the location of the falling object based on the analysis result; Formulate the shortest inspection route based on the location of the falling object and the location of each inspection terminal (6); The inspection route is sent to the inspection terminal (6).

6. A building high-altitude falling object intelligent detection and warning system applied to the building high-altitude falling object intelligent detection method according to any one of claims 1 to 5, characterized in that: It comprises a detection unit (1), a control unit (2) and an alarm unit (3); the detection unit (1) and the alarm unit (3) are both electrically connected to the control unit (2); The detection unit (1) is used to capture a picture of the building; The control unit (2) is used to obtain a picture taken by the detection unit; determine whether there is a falling object from a high altitude based on the picture; if so, perform image analysis on the picture to obtain an analysis result; and control the alarm unit to issue an early warning based on the analysis result; The alarm unit (3) is used to issue an early warning; The control unit (2) is specifically used to perform image analysis on the captured image, and the analysis results obtained include: Obtaining the background image captured by the detection unit (1); Performing image comparison between the background image and the captured image; Determining the position of the falling object within the captured image; Amplifying the captured image based on the position of the falling object to obtain an amplified image; Determining the type of the falling object based on the magnified image; Determining whether the falling object causes harm to the human body based on the type of the falling object; If yes, the type of the falling object and the judgment result are used as the analysis result, wherein the judgment result includes whether the falling object causes harm to the human body; The control unit (2) is specifically configured to, after determining the type of the high-altitude falling object based on the magnified image, further include: Obtaining support information of the falling object, wherein the support information includes cracks, damage, and support structure of the support; Determining the stability of the falling object based on the support information; Obtaining weather information of the environment where the high-altitude falling object is located; Determining a first probability of the falling object from high altitude based on the stability and the weather information; Taking the first probability of falling as an analysis result; The control unit (2) is specifically used to perform image analysis on the captured image, and obtaining the analysis result further includes: Acquire a historical shooting picture of the current detection unit based on the shooting picture; Counting the number of high-altitude falling objects based on the historical footage; Calculating a second probability of a high-altitude falling object phenomenon occurring in the area where the current detection unit is located based on the number of occurrences of the high-altitude falling object phenomenon; determining a danger level of the area where the detection unit is currently located based on the first probability and the second probability; The risk level is used as the analysis result.

7. The intelligent detection and early warning system for falling objects from high altitude buildings according to claim 6 is characterized in that: The system further comprises a communication unit (4) and an interception unit (5); the control unit (2) is wirelessly connected to the interception unit (5) via the communication unit (4); and further comprises: The inspection terminal (6) is wirelessly connected to the control unit and is used to receive the inspection route.

Images