High-rise object falling detection system and high-rise object falling detection method based on multi-source data fusion

Abstract

A high-rise object falling detection system based on multi-source data fusion, which relates to the technical field of high-rise object falling detection. The system comprises: acceleration sensors (11), sound sensors (12), a processor (13) and an alarm module (14), wherein the acceleration sensors (11) monitor an acceleration change of a target falling object, so as to preliminarily determine an object falling event; when the object falling event has been preliminarily determined, the sound sensors (12) monitor sound signals generated by the target falling object, so as to implement sound source positioning; the processor (13) receives data from the sensors, and reverifies the object falling event by means of a data fusion algorithm, so as to avoid erroneous determination; and the alarm module (14) generates alarm information, which comprises a falling time, an object type and a falling position, so as to raise an alarm. The problem in the prior art of the accuracy of performing object falling detection by means of a single-type sensor being relatively low is solved. Further disclosed is a high-rise object falling detection method.

Description

Multi-source data fusion high-altitude falling object detection system and method

[0001] This application claims priority to Chinese Patent Application No. 202411553262.0, filed on November 1, 2024, entitled "Multi-source data fusion system and method for detecting falling objects from heights", the entire contents of which are incorporated herein by reference. Technical Field

[0002] This invention relates to the field of falling object detection technology, specifically to a multi-source data fusion falling object detection system, method, device, and computer-readable storage medium. Background Technology

[0003] With the rapid development of society and economy, incidents of objects being thrown from high-rise buildings occur frequently, seriously threatening public safety and personal safety, and posing a great threat to society.

[0004] Currently, video surveillance systems are commonly used to detect falling objects from heights. These systems allow for real-time monitoring of such incidents and provide post-incident evidence. However, video surveillance systems have several drawbacks: First, they have blind spots, failing to detect falling objects in areas without supervision or coverage. Second, they can only provide post-incident evidence, exhibiting significant time lag. Third, their detection accuracy depends on the quality of the video feed; in areas with poor video quality, detection accuracy is low.

[0005] To address the aforementioned shortcomings, existing technical literature [Publication No. CN110911164A] proposes a method for detecting falling objects from heights based on video surveillance. This method monitors the occurrence of falling object events in real time through a video surveillance system, determines the occurrence of a falling object event by analyzing changes in the number of objects in the video, and determines the fall time, location, and duration by analyzing the object's trajectory. However, this method relies on the quality of the video surveillance; for areas with poor video surveillance quality, the detection accuracy of falling object events is low. Furthermore, the detection accuracy of this method depends on the temporal resolution of the video surveillance system; for areas with low temporal resolution of the video surveillance system, the detection accuracy of falling object events is also low. Summary of the Invention

[0006] The main objective of this application is to provide a high-altitude falling object detection system, method, device, and computer-readable storage medium that integrates multi-source data, so as to at least solve the problem of low accuracy in object falling detection using a single type of sensor in the prior art.

[0007] To achieve the above objectives, according to one aspect of this application, a high-altitude falling object detection system based on multi-source data fusion is provided, comprising: an accelerometer for monitoring changes in the acceleration of a falling object and preliminarily determining whether a falling object event exists; a sound sensor, communicatively connected to the accelerometer, for monitoring the sound signal generated by the falling object when a falling object event is detected, to achieve sound source localization and the type of the falling object; and a processor, communicatively connected to the accelerometer and the sound sensor, for receiving sensor data and re-verifying the falling object event through a data fusion algorithm to avoid misjudgment.

[0008] Optionally, the high-altitude falling object detection system also includes: an alarm module, which is connected in communication with the processor, and issues an alarm message to prompt personnel to handle the situation when the processor determines that a falling object event has occurred. The alarm message includes at least the time of fall, the type of the falling object, and the location of fall; and a power module, which is electrically connected to the acceleration sensor, the sound sensor, the processor, and the alarm module, and is used to supply power to the acceleration sensor, the sound sensor, the processor, and the alarm module.

[0009] According to another aspect of this application, a high-altitude falling object detection method using multi-source data fusion is provided. This method is used in a high-altitude falling object detection system and includes: controlling an accelerometer to monitor the acceleration change of a falling object to obtain a first target signal, and determining whether a falling object event exists based on the first target signal; if the accelerometer determines that a falling object event exists, controlling each sound sensor to monitor the sound signal of the falling object to obtain multiple second target signals; controlling a processor to receive the first and second target signals, and performing data fusion based on the first and second target signals to obtain a third target signal, and verifying the first target signal based on the third target signal.

[0010] Optionally, determining whether a falling object event exists based on the first target signal includes: calculating the product of gravitational acceleration and a preset ratio to obtain a first threshold; and determining that a falling object event exists if the acceleration of the falling object is greater than or equal to the first threshold.

[0011] Optionally, a third target signal is obtained by data fusion based on the first target signal and the second target signal, including: determining the falling acceleration of the target falling object based on the first target signal; determining the target sound pressure level of the sound generated by the target falling object based on the second target signal; obtaining a first preset weight and a second preset weight, wherein the first preset weight and the second preset weight are weight parameters for the falling acceleration and the target sound pressure level, respectively; performing a weighted calculation based on the falling acceleration, the target sound pressure level, the first preset weight, and the second preset weight to obtain the third target signal; and determining that a falling object event exists if the third target signal is greater than or equal to a second threshold.

[0012] Optionally, after controlling each sound sensor to monitor the sound signal of the falling target object and obtaining multiple second target signals, the method further includes: determining the time when the sound sensor corresponding to each second target signal receives the second target signal to obtain multiple target times; determining the time difference between each sound sensor receiving the second target signal based on the target times to obtain target duration; determining the distance difference between each sound sensor and the falling target object based on each target duration to obtain a first target distance; acquiring the positional deviation between each sound sensor to obtain a second target distance; locating the falling target object based on the first target distance and the second target distance to obtain target position information; and determining the frequency of the sound generated by the falling target object according to the second target signals to obtain a target frequency.

[0013] Optionally, after controlling the accelerometer to monitor the acceleration change of the target falling object and obtain the first target signal, the method further includes: integrating based on the first target signal to determine the falling speed of the target falling object.

[0014] Optionally, after verifying the first target signal based on the third target signal, the method further includes: determining the type of the falling object as a heavy object when the target frequency is greater than or equal to a third threshold and less than or equal to a fourth threshold; determining the type of the falling object as a light object when the target frequency is greater than a fourth threshold and less than or equal to a fifth threshold; determining the falling height of the falling object based on the target location information; determining the falling time of the falling object based on the falling height and falling speed; determining the falling position of the falling object based on the target location information, falling speed, and falling time; and generating alarm information based on the type of the falling object, the falling time, and the falling position.

[0015] According to another aspect of this application, a high-altitude falling object detection device with multi-source data fusion is provided. The high-altitude falling object detection device is used in the above-mentioned high-altitude falling object detection system, comprising: a first control unit, configured to control an accelerometer to monitor the acceleration change of a target falling object, obtain a first target signal, and determine whether a falling object event exists based on the first target signal; a second control unit, configured to control each sound sensor to monitor the sound signal of the target falling object when the accelerometer determines that a falling object event exists, and obtain multiple second target signals; and a third control unit, configured to control a processor to receive the first target signal and the second target signal, perform data fusion based on the first target signal and the second target signal to obtain a third target signal, and verify the first target signal based on the third target signal.

[0016] According to another aspect of this application, a computer-readable storage medium is provided, the computer-readable storage medium including a stored program, wherein, when the program is executed, it controls the device on which the computer-readable storage medium is located to perform any one of the methods.

[0017] By combining data from an accelerometer and a sound sensor, the accuracy and reliability of detecting falling objects from heights are improved. The accelerometer can quickly respond to falling object events, while the sound sensor can locate the sound source. The combination of the two not only allows for timely detection of falling objects but also accurately determines their type and location, effectively avoiding misjudgments that may occur with a single sensor. Furthermore, verification through a data fusion algorithm further improves the detection accuracy, providing strong technical support for the safety management of falling objects from heights. In practical applications, this system can significantly improve public safety and reduce potential risks and losses caused by falling objects. The above embodiments solve the problem of low accuracy in object falling detection using a single type of sensor in existing technologies. Attached Figure Description

[0018] Figure 1 shows a structural block diagram of a high-altitude falling object detection system based on multi-source data fusion provided in an embodiment of this application;

[0019] Figure 2 shows a flowchart of a multi-source data fusion method for detecting falling objects from heights according to an embodiment of this application;

[0020] Figure 3 shows a structural block diagram of a high-altitude falling object detection device based on an embodiment of this application, which incorporates multi-source data fusion.

[0021] The above-mentioned figures include the following reference numerals: 11, accelerometer; 12, sound sensor; 13, processor; 14, alarm module; 15, power supply module. Detailed Implementation

[0022] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0023] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0024] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate for the embodiments of this application described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0025] As described in the background section, existing technologies for detecting falling objects from heights using video surveillance systems rely on the quality of the video surveillance. In areas with poor video surveillance quality, the detection accuracy of falling object events is low. To address the problem of low accuracy in detecting falling objects using a single type of sensor in existing technologies, embodiments of this application provide a multi-source data fusion system, method, apparatus, and computer-readable storage medium for detecting falling objects from heights.

[0026] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0027] This embodiment provides a high-altitude falling object detection system based on multi-source data fusion, as shown in Figure 1. The system includes: an accelerometer 11, a sound sensor 12, a processor 13, an alarm module 14, and a power supply module 15, wherein:

[0028] Accelerometer 11 is used to monitor the acceleration changes of a falling object and to make a preliminary judgment on whether a falling object event has occurred.

[0029] The sound sensor 12 is communicatively connected to the aforementioned accelerometer and is used to monitor the sound signal generated by the falling object when the aforementioned falling object event is detected, so as to locate the sound source and determine the type of the falling object.

[0030] The processor 13 is connected to the accelerometer and the sound sensor to receive sensor data and re-verify the falling object event through a data fusion algorithm to avoid false judgment.

[0031] Alarm module 14 is connected to the processor. When the processor determines that a falling object event has occurred, it issues an alarm message to prompt the staff to handle the situation. The alarm message includes at least the time of the fall, the type of the falling object, and the location of the fall.

[0032] The power supply module 15 is electrically connected to the acceleration sensor, sound sensor, processor, and alarm module, and is used to supply power to the acceleration sensor, sound sensor, processor, and alarm module.

[0033] Based on the above embodiments, in specific implementation, the installation of the sensor and data acquisition include:

[0034] 1.1 Arrangement of acceleration sensors: Multiple acceleration sensors 11 are installed at key locations on the top, bottom and all sides of the transmission tower to ensure full coverage of all facades and height range of the tower and capture the acceleration changes of the target object when it falls.

[0035] 1.2 Data Acquisition and Preliminary Time Diagnosis by the Accelerometer Sensor: After installation, the accelerometer sensor 11 continuously monitors acceleration changes within the target area. If an abnormal increase in acceleration is detected, exceeding a preset value, a preliminary judgment is made that a falling object may have occurred. The accelerometer sensor 11 synchronously transmits this judgment result to the processor and the sound sensor 12.

[0036] 1.3 Arrangement of sound sensors: Sound sensors are distributed around the transmission tower to form a three-dimensional positioning network.

[0037] 1.4 Sound Source Localization and Object Type Recognition: After the accelerometer initially determines that a falling object is involved, the sound sensor array begins to work, capturing the sound signal of the falling object. A time-difference time-of-arrival (TDOA) localization algorithm is used to calculate the time difference Δt between the sound signals received by multiple sensors. ij Through the speed of sound propagation v sound Calculate the distance difference Δd ij This allows us to determine the three-dimensional coordinates (x, y) of the sound source. s y s , z sFurthermore, the type of object is determined based on the frequency and sound pressure level of the sound, such as lightweight objects (e.g., plastic bags) and heavy objects (e.g., iron tools).

[0038] Based on the data acquired from the aforementioned sensors, the processor further performs high-altitude falling object verification using a data fusion algorithm, including:

[0039] 2.1 Data Reception and Preprocessing: The processor receives raw data from the accelerometer and sound sensor, and removes noise interference through signal processing algorithms such as filtering to ensure the purity and reliability of the data.

[0040] 2.2 Data Fusion: The processor performs comprehensive analysis on the preprocessed acceleration data and sound data to correct the initial judgment of a falling object incident. In an optional implementation, the acceleration sensor data a(t) and the sound sensor data Lp (i.e., the aforementioned sound pressure level) can be assigned corresponding weights according to the actual situation, and data fusion can be performed by weighted averaging to ensure balanced use of data from both sensors.

[0041] 2.3 Event Verification and Alarm: Based on the fused data, a preset value is compared. If the fused data exceeds the preset value, a falling object event is confirmed. Furthermore, the processor sends relevant time information (time, location, and object type, etc.) to a designated terminal via a communication connection, while simultaneously recording the event data within the system for subsequent statistical analysis.

[0042] Based on the alarm from the aforementioned processor, the alarm module further performs the following alarm functions:

[0043] 3.1 Alarm Module: The alarm module of this application is equipped with one or more alarm signal generators and a wireless communication device. The alarm signal generator is used to generate alarm sound or light signals, and the wireless communication device supports 4G, 5G and satellite communication to ensure fast and reliable alarm activation.

[0044] 3.2 Alarm Logic: The alarm module is connected to the processor. After the processor determines that a falling object event has occurred through a data fusion algorithm, it sends a signal to the alarm module, including the falling time, the type of the target falling object, and the falling location, to drive the alarm module to sound an alarm.

[0045] Based on the above embodiments, in an optional real-time mode, the power module of this application comprises: a solar panel, a battery management system (BMS), and a backup battery. The solar panel is used to collect energy under sunlight conditions, converting solar energy into electrical energy. The BMS is responsible for monitoring the power status, ensuring efficient energy utilization and the health of the battery. The backup battery provides stable power at night or in cloudy / rainy weather, ensuring continuous system operation.

[0046] In the above embodiments, to ensure the accuracy of the alarm, in an optional implementation, the system can perform further algorithm optimization during use, including:

[0047] 4.1 Sensitivity Adjustment: To adapt to different environmental conditions, the system has a sensitivity adjustment function. In noisy environments, the threshold of the sound sensor can be increased to reduce false alarms; in areas with a high risk of falling objects, the threshold of the accelerometer can be appropriately lowered to ensure that small falling objects can be detected in a timely manner.

[0048] 4.2 False Alarm Control: By continuously optimizing the data fusion algorithm and sensor threshold settings, the system can effectively control the false alarm rate and avoid unnecessary alarms. For example, machine learning models are used to predict the ambient noise level and dynamically adjust the sensor thresholds to improve detection accuracy in noisy environments.

[0049] The above embodiments, combining data from accelerometers and sound sensors, improve the accuracy and reliability of falling object detection from heights. Accelerometers can quickly respond to falling object events, while sound sensors can locate the sound source. The combination of both not only allows for timely detection of falling objects but also accurately determines their type and location, effectively avoiding misjudgments that might occur with a single sensor. Furthermore, data fusion algorithms further enhance detection accuracy, providing strong technical support for the safety management of falling objects from heights. In practical applications, this system can significantly improve public safety and reduce potential risks and losses caused by falling objects. These embodiments address the problem of low accuracy in object falling detection using a single type of sensor in existing technologies.

[0050] This embodiment provides a high-altitude falling object detection method that integrates multi-source data and runs on a mobile terminal, computer terminal, or similar computing device. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0051] Figure 2 is a flowchart of a high-altitude falling object detection method based on multi-source data fusion according to an embodiment of this application. As shown in Figure 2, the method includes the following steps:

[0052] Step S201: Control the above-mentioned accelerometer to monitor the acceleration change of the target falling object, obtain the first target signal, and determine whether there is a falling object event based on the above-mentioned first target signal;

[0053] In practical implementation, firstly, upon system startup, the processor activates the accelerometer, putting it into real-time monitoring mode. Then, the accelerometer acquires data, continuously measuring acceleration changes within the target area to obtain real-time acceleration data a(t). Simultaneously, based on this real-time acceleration data (the aforementioned first target signal), and by comparing it with a preset threshold a... thresha The system compares the data to determine if a falling object event has occurred. Specifically, if the real-time acceleration data exceeds the preset threshold, a falling object event is confirmed, and the data is transmitted to the processor and sound sensor.

[0054] Step S202: When the acceleration sensor determines that the falling object event exists, control each of the sound sensors to monitor the sound signal of the falling object and obtain multiple second target signals.

[0055] In practical implementation, firstly, when the accelerometer initially determines the presence of a falling object based on acceleration data, it controls the sound sensor array to start and capture the sound signal of the falling object; then, each sensor S1, S2, ..., S in the sound sensor array... n The recording of the sound signal generated by the falling object begins, acquiring multiple time-domain sound signal data Lp(t) (i.e., the aforementioned second target signal). Based on these time-domain sound signals, sound source localization and object type identification are performed. Time-frequency analysis is then conducted based on the second target signal to extract the frequency and intensity features of the sound. The time difference Δt between the sound signals received by different sensors is analyzed using the Time Difference of Origin (TDOA) localization algorithm. ij Calculate the distance difference Δd between the sound sources. ij And determine the three-dimensional coordinates (x, y) of the sound source. s y s , z s This allows for precise location of the falling object. Simultaneously, based on the frequency and intensity characteristics of the sound, the type of falling object is identified, and the results are transmitted to the aforementioned processor.

[0056] Step S203: Control the processor to receive the first target signal and the second target signal, perform data fusion based on the first target signal and the second target signal to obtain a third target signal, and verify the first target signal based on the third target signal.

[0057] Specifically, the processor removes interference by preprocessing (e.g., filtering) the first target signal a(t) and multiple second target signals Lp(t) from the sound sensor to ensure data purity. Based on the preprocessed data, a data fusion algorithm is run to weightedly fuse the first target signal a(t) and the second target signals Lp(t) to form a third target signal F(t), which integrates the motion state and sound characteristics of the falling object. The processor then confirms and verifies the falling object event based on this third target signal, using a combination of the third target signal F(t) and a comprehensive threshold F. thresh Compare and verify the existence of the falling object incident again. F thresh The setting needs to comprehensively consider the acceleration, background noise level of the sound signal, and event characteristics. When F(t) ≥ F thresh At that time, the system finally confirmed the falling object incident and issued an alarm message through the alarm module.

[0058] In the above embodiments, to ensure safety, in one optional implementation, after the falling object incident is confirmed, the system issues an alarm through the alarm module and also sends detailed event information (including time, location, and type of falling object) to a remote terminal through the wireless communication module, ensuring that staff can respond quickly and take appropriate measures.

[0059] In this embodiment, firstly, the aforementioned accelerometer is controlled to monitor the acceleration changes of a falling object, obtaining a first target signal, and determining whether a falling object event exists based on the first target signal. Then, if the accelerometer determines that a falling object event exists, each of the aforementioned sound sensors is controlled to monitor the sound signal of the falling object, obtaining multiple second target signals. Finally, the aforementioned processor is controlled to receive the first and second target signals, and performs data fusion based on the first and second target signals to obtain a third target signal, which is then used to verify the first target signal. This application detects falling events by controlling an accelerometer, locates falling objects by controlling sound sensors, and verifies the authenticity of the falling event by controlling the processor to perform data fusion on the monitoring signals from the accelerometer and sound sensors. This ensures the accuracy of falling event detection and solves the problem of low accuracy in object falling detection using a single type of sensor in the prior art.

[0060] In order to make a preliminary diagnosis of a fall incident, in one optional implementation, step S201 above includes:

[0061] Step S2011: Calculate the product of gravitational acceleration and a preset ratio to obtain the first threshold.

[0062] In specific implementation, the above-mentioned first threshold calculation includes, after the processor receives the acceleration data, calculating the product of the gravitational acceleration g (approximately 9.8 m / s²) and a preset ratio α to obtain the first threshold α. thresh = g × α. Wherein, α is set according to the on-site environment and object type, generally between 0.8 and 1.2, to adapt to the detection needs of different falling object events.

[0063] Step S2012: If the acceleration of the falling object is greater than or equal to the first threshold, it is determined that the falling object event exists.

[0064] In practical implementation, the processor compares the real-time collected acceleration data a(t) with the first threshold a. thresh Compare the results. Diagnose the falling object incident based on the judgment result; if a(t) ≥ a thresh If the acceleration change of the target object exceeds the preset criteria for judging a falling object event, the processor will immediately determine that a falling object event has occurred.

[0065] In the above embodiments, to ensure the accuracy of the diagnosis, in an optional implementation, this application sets the above-mentioned acceleration sensor for environmental adaptation, including: dynamic threshold adjustment: the system has a dynamic threshold adjustment function, which can adjust the first threshold a according to environmental changes in the monitored area (such as wind speed, temperature, etc.). thresh This ensures that the system can accurately detect falling objects from heights under different conditions; adaptive event detection: by calculating and adjusting the first threshold in real time, the system can adaptively monitor various possible falling events, whether it is a slow fall of a light object or a rapid fall of a heavy object, it can be detected in time.

[0066] In order to perform multi-source data fusion, in one optional implementation, step S203 above includes:

[0067] Step S2031: Based on the first target signal, determine the falling acceleration of the falling object.

[0068] Specifically, the fall acceleration calculation is as follows: When the system initially determines that there may be a falling object from a height, the processor calculates the real-time fall acceleration of the falling object based on the first target signal a(t) collected by the accelerometer. Considering that the acceleration of an object in free fall should be close to the gravitational acceleration g, the processor calculates the instantaneous fall acceleration of the object by monitoring the changes in a(t) in real time.

[0069] Step S2032: Based on the second target signal, determine the target sound pressure level of the sound generated by the falling object.

[0070] Specifically, the target sound pressure level is acquired as follows: Simultaneously, upon detecting a falling object event, the sound sensor monitors and records a second target signal, i.e., the sound signal generated when the object falls. The processor performs time-frequency analysis on the sound signal, extracts the sound intensity features, and calculates the target sound pressure level Lp of the sound generated by the falling object.

[0071] Step S2033: Obtain the first preset weight and the second preset weight, wherein the first preset weight and the second preset weight are the weight parameters of the falling acceleration and the target sound pressure level, respectively.

[0072] Specifically, weight parameter acquisition: The processor obtains the first preset weight w based on preset weight parameters. a Second preset weight w s These two weighting parameters reflect the importance of fall acceleration and target sound pressure level in event assessment, and are determined through system training or on-site adjustments to balance the contributions of the two sensor data.

[0073] Step S2034: Based on the above-mentioned fall acceleration, the above-mentioned target sound pressure level, the above-mentioned first preset weight and the above-mentioned second preset weight, a weighted calculation is performed to obtain the above-mentioned third target signal;

[0074] Specifically, the processor is based on the fall acceleration, the target sound pressure level, and a first preset weight w. a Second preset weight w s We then perform a weighted calculation. The specific formula is as follows: F(t) = w a ×a(t)+w s ×Lp(t), where F(t) is the third target signal, which is a real-time fused signal that integrates the fall acceleration and sound intensity.

[0075] Step S2035: If the third target signal is greater than or equal to the second threshold, it is determined that the aforementioned falling object event exists.

[0076] Specifically, first, the processor sets a second threshold F. thresh This serves as a comprehensive criterion for determining whether an object has fallen from the sky. thresh The settings are based on the system's false alarm and false negative rates under different environments, and were optimized through a large training dataset to ensure the accuracy and reliability of the system in detecting falling objects. The processor then calculates the third target signal F(t) in real time and modifies it with the second threshold F. thresh Compare them. When F(t) ≥ F thresh At that point, the system finally determined that a falling object incident had occurred.

[0077] In the above embodiments, to ensure the accuracy of processor verification, in an optional implementation, this application sets the processor to perform environmental adaptation, including: environmental adaptability adjustment: the system has a dynamic adjustment mechanism for weight parameters, which can adjust the first preset weight w in real time according to factors such as environmental noise level and falling object risk area. a Second preset weight w s To adapt to the falling object detection needs in different scenarios. System optimization and false alarm control: By continuously optimizing the weighted calculation model, the system can effectively reduce false alarms and false negatives. For example, in noisy environments, w can be appropriately reduced. s Increase w a The proportion of [something] is adjusted to ensure that interference with the sound signal does not lead to false alarms; in areas with a high risk of falling objects, [something] can be increased. s This allows for more sensitive capture of sound signals, improving detection accuracy.

[0078] In an optional implementation for sound source localization and falling object classification, after controlling each of the aforementioned sound sensors to monitor the sound signal of the target falling object and obtaining multiple second target signals, the method further includes:

[0079] Step S301: Determine the time when the sound sensor receives the second target signal corresponding to each of the above-mentioned second target signals, and obtain multiple target times;

[0080] In practice, the system immediately controls each sound sensor to begin monitoring the sound signal generated by the falling target object the instant an acceleration anomaly is detected. The processor records the precise time t when each sound sensor receives the second target signal. i (i = 1, 2, ..., n), where n is the number of sound sensors.

[0081] Step S302: Based on the target time, determine the time difference between the reception of the second target signal by each of the aforementioned sound sensors, and obtain the target duration;

[0082] In practical implementation, based on the recorded target time t i Calculate the time difference Δt between the received signals of any two sound sensors. ij =t i -t j Here, i and j represent two different sets of sensors. The time difference reflects the difference in the time it takes for the sound wave to reach the different sensors, and is the core parameter for sound source localization.

[0083] Step S303: Based on the duration of each of the above targets, determine the distance difference between each of the above sound sensors and the falling target to obtain the first target distance;

[0084] In practice, the processor obtains the speed of sound propagation in the air, vsound (approximately 343 m / s), and utilizes the time difference Δt ij Calculate the distance difference Δd between each sound sensor and the sound source. ij =vsound×Δt ij By calculating the distance difference, the relative position of the falling object to each sensor can be further determined.

[0085] Step S304: Obtain the positional deviation between each of the aforementioned sound sensors to obtain the second target distance; based on the aforementioned first target distance and the aforementioned second target distance, locate the aforementioned falling object to obtain target position information;

[0086] Specifically, the processor will calculate the distance difference Δd. ij By combining the positional deviation between sensors (the known distance between sensors), the precise position of the falling object is determined using nonlinear solution methods (such as least squares or iterative optimization algorithms). This process utilizes mathematical models and algorithms to determine the three-dimensional coordinate position (x, y, z) of the falling object. s y s , z s This enables spatial location tracking of falling objects.

[0087] Step S305: Determine the frequency of the sound generated by the falling target object based on the second target signal to obtain the target frequency.

[0088] Specifically, the sound sensor not only monitors the arrival time of the sound signal but also records its frequency characteristics. Based on the second target signal Lp(t), the processor converts the time-domain signal into a frequency-domain signal through Fourier transform to obtain the spectral characteristics of the sound.

[0089] In the above embodiments, to ensure the accuracy of data analysis, in an optional implementation, this application sets continuously adjustable parameters for the time difference positioning algorithm and a frequency analysis model for the sound signal to adapt to different monitoring environments and noise levels, thereby improving the system's detection accuracy and robustness. By setting reasonable sound source localization thresholds and frequency analysis ranges, the system can significantly reduce false alarms caused by environmental noise or other interference, ensuring that each alarm is based on accurate falling object event data. Through sound source localization technology, the system can accurately determine the location of falling objects, making it suitable for high-rise residential areas, commercial complexes, and other locations requiring precise location information, providing important evidence for timely removal of falling objects and prevention of secondary injuries.

[0090] To assess the impact of a falling object, in one optional implementation, after controlling the aforementioned accelerometer to monitor the acceleration changes of the falling target and obtain a first target signal, the method further includes:

[0091] Step S401: Integrate based on the first target signal to determine the falling speed of the falling object.

[0092] In practice, when the system initially determines that a falling object event may occur, the processor begins to perform integration on the first target signal a(t). The integration formula is: v(t)=∫a(t)dt+v0, where v(t) is the instantaneous velocity of the object at time t, and v0 is the initial velocity (in the case of free fall, the initial velocity is usually 0).

[0093] Based on the above embodiments, the processor can also compare the calculated instantaneous velocity v(t) with a preset velocity threshold v thresh Compare them. If v(t) ≥ v thresh This further confirms the falling object incident and incorporates speed information into the event analysis, providing support for subsequent sound source localization and object type identification. Specifically, to ensure the accuracy and reliability of the event judgment, the processor combines v(t) with the falling object location and sound signal intensity features obtained from the time difference localization algorithm to perform multi-dimensional verification of the falling object incident.

[0094] In the above embodiments, to ensure the accuracy of data analysis, in an optional implementation, this application sets up a system that can adjust its detection and response strategies by calculating the falling velocity v(t) in real time. For example, when the falling velocity is abnormally high, the system can automatically raise the alarm level to ensure that emergencies are handled promptly. The falling velocity information v(t) is not only used for real-time monitoring and event verification, but also for generating event reports. The report contains the velocity data of the falling object, helping managers assess the impact force and potential hazards of the falling object, providing a quantitative basis for safety management and risk prevention. The calculation of the falling velocity helps to further determine the degree of danger of the falling object, and is applicable to infrastructure such as bridges and tunnels that require assessment of the impact force of falling objects, providing data support for maintaining structural safety.

[0095] To facilitate the handling of fall incidents, in an optional implementation, after verifying the first target signal based on the third target signal, the method further includes:

[0096] Step S501: If the target frequency is greater than or equal to the third threshold and less than or equal to the fourth threshold, determine that the type of the target falling object is a heavy object; if the target frequency is greater than the fourth threshold and less than or equal to the fifth threshold, determine that the type of the target falling object is a light object.

[0097] In practice, the processor sets a third threshold F3 and a fourth threshold F4 to distinguish the frequency characteristics of heavy and light objects. These thresholds are set according to a preset acoustic model for object types, typically F3 = 20Hz and F4 = 10kHz, to accommodate the falling frequency range of different objects. The processor performs frequency analysis on the second target signal collected by the sound sensor, extracting the target frequency f. By comparing f with F3 and F4, the type of falling object is determined. If the target frequency f is greater than or equal to F3 and less than or equal to F4, the processor identifies the falling object as a heavy object; if f is greater than F4 and less than or equal to the fifth threshold F5 (e.g., F5 = 20kHz), the target object is classified as a light object. This process ensures that the system can intelligently identify the object type based on the sound characteristics of the falling object, providing crucial information for event handling.

[0098] Step S502: Determine the falling height of the target falling object based on the target location information; determine the falling time of the target falling object based on the falling height and the falling speed; and determine the falling position of the target falling object based on the target location information, the falling speed, and the falling time.

[0099] Specifically, the processor is based on target location information (x s y s , z s Determine the falling height h of the target object. The falling height h can be calculated from the vertical coordinate difference in the target's position information, i.e., h = Hz. s Here, H represents the height of the pole or monitoring area. Using the free-fall motion formula and the known fall height h, the processor can deduce the fall time t. By calculating the fall time, the system can more accurately determine the entire process of an object's fall. The processor combines the fall velocity v(t), fall time t, and target position information to perform three-dimensional spatial positioning, determining the precise fall location of the target object. This process ensures that the system can provide detailed spatial information about the falling object event, supporting subsequent event analysis and processing.

[0100] Step S503: An alarm message is generated based on the type of the falling object, the falling time, and the falling location.

[0101] In practice, the processor constructs alarm information based on information such as the type of falling object, the time of fall, and the location of fall, including key parameters such as the time of fall, object type, fall height, and precise fall location.

[0102] In the above embodiments, the system automatically records detailed information for each falling object incident, including alarm information, the type of falling object, the time and location of the fall, and generates an incident report. This data is used not only for immediate response but also for subsequent incident analysis, helping managers identify falling object risk trends and formulate effective risk prevention strategies. This type identification and detailed information recording function enables the system to provide comprehensive information after detecting a falling object, making it suitable for scenarios such as high-rise building management and urban security monitoring that require detailed accident reports, providing strong evidence for accident investigation and liability determination.

[0103] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0104] This application also provides a high-altitude falling object detection device based on multi-source data fusion. It should be noted that this device can be used to execute the high-altitude falling object detection method based on multi-source data fusion provided in this application. This device is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0105] The following describes the high-altitude falling object detection device with multi-source data fusion provided in the embodiments of this application.

[0106] Figure 3 is a structural block diagram of a high-altitude falling object detection device based on a multi-source data fusion embodiment of this application. As shown in Figure 3, the device includes:

[0107] The first control unit 10 is used to control the aforementioned acceleration sensor to monitor the acceleration change of the target falling object, obtain a first target signal, and determine whether a falling object event exists based on the aforementioned first target signal;

[0108] In practical implementation, firstly, upon system startup, the processor activates the accelerometer, putting it into real-time monitoring mode. Then, the accelerometer acquires data, continuously measuring acceleration changes within the target area to obtain real-time acceleration data a(t). Simultaneously, based on this real-time acceleration data (the aforementioned first target signal), and by comparing it with a preset threshold a... thresha The system compares the data to determine if a falling object event has occurred. Specifically, if the real-time acceleration data exceeds the preset threshold, a falling object event is confirmed, and the data is transmitted to the processor and sound sensor.

[0109] The second control unit 20 is used to control each of the above-mentioned sound sensors to monitor the sound signal of the target falling object when the above-mentioned acceleration sensor determines that the above-mentioned falling object event exists, and to obtain multiple second target signals;

[0110] In practical implementation, firstly, when the accelerometer initially determines the presence of a falling object based on acceleration data, it controls the sound sensor array to start and capture the sound signal of the falling object; then, each sensor S1, S2, ..., S in the sound sensor array... n The recording of the sound signal generated by the falling object begins, acquiring multiple time-domain sound signal data Lp(t) (i.e., the aforementioned second target signal). Based on these time-domain sound signals, sound source localization and object type identification are performed. Time-frequency analysis is then conducted based on the second target signal to extract the frequency and intensity features of the sound. The time difference Δt between the sound signals received by different sensors is analyzed using the Time Difference of Origin (TDOA) localization algorithm. ij Calculate the distance difference Δd between the sound sources. ij And determine the three-dimensional coordinates (x, y) of the sound source. s y s , z s This allows for precise location of the falling object. Simultaneously, based on the frequency and intensity characteristics of the sound, the type of falling object is identified, and the results are transmitted to the aforementioned processor.

[0111] The third control unit 30 is used to control the processor to receive the first target signal and the second target signal, perform data fusion based on the first target signal and the second target signal to obtain a third target signal, and verify the first target signal based on the third target signal.

[0112] Specifically, the processor removes interference by preprocessing (e.g., filtering) the first target signal a(t) and multiple second target signals Lp(t) from the sound sensor to ensure data purity. Based on the preprocessed data, a data fusion algorithm is run to weightedly fuse the first target signal a(t) and the second target signals Lp(t) to form a third target signal F(t), which integrates the motion state and sound characteristics of the falling object. The processor then confirms and verifies the falling object event based on this third target signal, using a combination of the third target signal F(t) and a comprehensive threshold F. thresh Compare and verify the existence of the falling object incident again. F thresh The setting needs to comprehensively consider the acceleration, background noise level of the sound signal, and event characteristics. When F(t) ≥ F thresh At that time, the system finally confirmed the falling object incident and issued an alarm message through the alarm module.

[0113] In the above embodiments, to ensure safety, in one optional implementation, after the falling object incident is confirmed, the system issues an alarm through the alarm module and also sends detailed event information (including time, location, and type of falling object) to a remote terminal through the wireless communication module, ensuring that staff can respond quickly and take appropriate measures.

[0114] In this embodiment, the first control unit controls the aforementioned accelerometer to monitor the acceleration changes of a falling object, obtaining a first target signal, and determines whether a falling object event exists based on the first target signal. When the accelerometer determines that a falling object event exists, the second control unit controls each of the aforementioned sound sensors to monitor the sound signal of the falling object, obtaining multiple second target signals. The third control unit controls the aforementioned processor to receive the first and second target signals, and performs data fusion based on the first and second target signals to obtain a third target signal, which is then used to verify the first target signal. This application detects falling events by controlling the accelerometer and locates the falling object by controlling the sound sensor. Furthermore, by controlling the processor to perform data fusion on the monitoring signals from the accelerometer and sound sensor to verify the authenticity of the falling event, the accuracy of falling event detection is ensured, solving the problem of low accuracy in object falling detection using a single type of sensor in the prior art.

[0115] In order to perform a preliminary diagnosis of a fall incident, in one optional implementation, the first control unit includes:

[0116] The first calculation module is used to calculate the product of gravitational acceleration and a preset ratio to obtain the first threshold.

[0117] In specific implementation, the above-mentioned first threshold calculation includes, after the processor receives the acceleration data, calculating the product of the gravitational acceleration g (approximately 9.8 m / s²) and a preset ratio α to obtain the first threshold α. thresh = g × α. Wherein, α is set according to the on-site environment and object type, generally between 0.8 and 1.2, to adapt to the detection needs of different falling object events.

[0118] The first determining module is used to determine the existence of the falling object event when the acceleration of the falling object is greater than or equal to the first threshold.

[0119] In practical implementation, the processor compares the real-time collected acceleration data a(t) with the first threshold a. thresh Compare the results. Diagnose the falling object incident based on the judgment result; if a(t) ≥ a threshIf the acceleration change of the target object exceeds the preset criteria for judging a falling object event, the processor will immediately determine that a falling object event has occurred.

[0120] In the above embodiments, to ensure the accuracy of the diagnosis, in an optional implementation, this application sets the above-mentioned acceleration sensor for environmental adaptation, including: dynamic threshold adjustment: the system has a dynamic threshold adjustment function, which can adjust the first threshold a according to environmental changes in the monitored area (such as wind speed, temperature, etc.). thresh This ensures that the system can accurately detect falling objects from heights under different conditions; adaptive event detection: by calculating and adjusting the first threshold in real time, the system can adaptively monitor various possible falling events, whether it is a slow fall of a light object or a rapid fall of a heavy object, it can be detected in time.

[0121] In an optional implementation, for multi-source data fusion, the third control unit includes:

[0122] The second determining module is used to determine the falling acceleration of the falling object based on the first target signal.

[0123] Specifically, the fall acceleration calculation is as follows: When the system initially determines that there may be a falling object from a height, the processor calculates the real-time fall acceleration of the falling object based on the first target signal a(t) collected by the accelerometer. Considering that the acceleration of an object in free fall should be close to the gravitational acceleration g, the processor calculates the instantaneous fall acceleration of the object by monitoring the changes in a(t) in real time.

[0124] The third determining module is used to determine the target sound pressure level of the sound generated by the falling object based on the second target signal mentioned above.

[0125] Specifically, the target sound pressure level is acquired as follows: Simultaneously, upon detecting a falling object event, the sound sensor monitors and records a second target signal, i.e., the sound signal generated when the object falls. The processor performs time-frequency analysis on the sound signal, extracts the sound intensity features, and calculates the target sound pressure level Lp of the sound generated by the falling object.

[0126] The first acquisition module is used to acquire a first preset weight and a second preset weight, wherein the first preset weight and the second preset weight are respectively weight parameters of the fall acceleration and the target sound pressure level;

[0127] Specifically, weight parameter acquisition: The processor obtains the first preset weight w based on preset weight parameters. a Second preset weight w s These two weighting parameters reflect the importance of fall acceleration and target sound pressure level in event assessment, and are determined through system training or on-site adjustments to balance the contributions of the two sensor data.

[0128] The second calculation module is used to perform weighted calculations based on the above-mentioned fall acceleration, the above-mentioned target sound pressure level, the above-mentioned first preset weight and the above-mentioned second preset weight to obtain the above-mentioned third target signal;

[0129] Specifically, the processor is based on the fall acceleration, the target sound pressure level, and a first preset weight w. a Second preset weight w s We then perform a weighted calculation. The specific formula is as follows: F(t) = w a ×a(t)+w s ×Lp(t), where F(t) is the third target signal, which is a real-time fused signal that integrates the fall acceleration and sound intensity.

[0130] The fourth determining module is used to determine the existence of the aforementioned falling object event when the aforementioned third target signal is greater than or equal to the second threshold.

[0131] Specifically, first, the processor sets a second threshold F. thresh This serves as a comprehensive criterion for determining whether an object has fallen from the sky. thresh The settings are based on the system's false alarm and false negative rates under different environments, and were optimized through a large training dataset to ensure the accuracy and reliability of the system in detecting falling objects. The processor then calculates the third target signal F(t) in real time and modifies it with the second threshold F. thresh Compare them. When F(t) ≥ F thresh At that point, the system finally determined that a falling object incident had occurred.

[0132] In the above embodiments, to ensure the accuracy of processor verification, in an optional implementation, this application sets the processor to perform environmental adaptation, including: environmental adaptability adjustment: the system has a dynamic adjustment mechanism for weight parameters, which can adjust the first preset weight w in real time according to factors such as environmental noise level and falling object risk area. a Second preset weight w s To adapt to the falling object detection needs in different scenarios. System optimization and false alarm control: By continuously optimizing the weighted calculation model, the system can effectively reduce false alarms and false negatives. For example, in noisy environments, w can be appropriately reduced. s Increase w a The proportion of [something] is adjusted to ensure that interference with the sound signal does not lead to false alarms; in areas with a high risk of falling objects, [something] can be increased. s This allows for more sensitive capture of sound signals, improving detection accuracy.

[0133] In an optional embodiment, for sound source localization and falling object classification, the above-mentioned device further includes:

[0134] The first determining unit is used to determine the time when the sound sensor corresponding to each of the above-mentioned sound sensors receives the sound signal of the falling target object after controlling each of the above-mentioned sound sensors to monitor the sound signal of the target falling object and obtain multiple second target signals, thereby obtaining multiple target times.

[0135] In practice, the system immediately controls each sound sensor to begin monitoring the sound signal generated by the falling target object the instant an acceleration anomaly is detected. The processor records the precise time t when each sound sensor receives the second target signal. i (i = 1, 2, ..., n), where n is the number of sound sensors.

[0136] The second determining unit is used to determine the time difference between the receipt of the second target signal by each of the aforementioned sound sensors based on the aforementioned target time, and to obtain the target duration.

[0137] In practical implementation, based on the recorded target time t i Calculate the time difference Δt between the received signals of any two sound sensors. ij =t i -t j Here, i and j represent two different sets of sensors. The time difference reflects the difference in the time it takes for the sound wave to reach the different sensors, and is the core parameter for sound source localization.

[0138] The third determining unit is used to determine the distance difference between each of the above-mentioned sound sensors and the target falling object based on the duration of each of the above-mentioned targets, and to obtain the first target distance;

[0139] In practice, the processor obtains the speed of sound propagation in the air, vsound (approximately 343 m / s), and utilizes the time difference Δt ij Calculate the distance difference Δd between each sound sensor and the sound source. ij =vsound×Δt ij By calculating the distance difference, the relative position of the falling object to each sensor can be further determined.

[0140] The fourth determining unit is used to obtain the positional deviation between each of the aforementioned sound sensors, obtain the second target distance, and locate the falling target based on the aforementioned first target distance and the aforementioned second target distance to obtain target position information;

[0141] Specifically, the processor will calculate the distance difference Δd. ij By combining the positional deviation between sensors (the known distance between sensors), the precise position of the falling object is determined using nonlinear solution methods (such as least squares or iterative optimization algorithms). This process utilizes mathematical models and algorithms to determine the three-dimensional coordinate position (x, y, z) of the falling object.s y s , z s This enables spatial location tracking of falling objects.

[0142] The fifth determining unit is used to determine the frequency of the sound generated by the falling target object based on the second target signal, and obtain the target frequency.

[0143] Specifically, the sound sensor not only monitors the arrival time of the sound signal but also records its frequency characteristics. Based on the second target signal Lp(t), the processor converts the time-domain signal into a frequency-domain signal through Fourier transform to obtain the spectral characteristics of the sound.

[0144] In the above embodiments, to ensure the accuracy of data analysis, in an optional implementation, this application sets continuously adjustable parameters for the time difference positioning algorithm and a frequency analysis model for the sound signal to adapt to different monitoring environments and noise levels, thereby improving the system's detection accuracy and robustness. By setting reasonable sound source localization thresholds and frequency analysis ranges, the system can significantly reduce false alarms caused by environmental noise or other interference, ensuring that each alarm is based on accurate falling object event data. Through sound source localization technology, the system can accurately determine the location of falling objects, making it suitable for high-rise residential areas, commercial complexes, and other locations requiring precise location information, providing important evidence for timely removal of falling objects and prevention of secondary injuries.

[0145] To assess the impact of falling objects from a height, in one alternative embodiment, the above-mentioned device further includes:

[0146] The first calculation unit is used to determine the falling speed of the target object by integrating the first target signal after controlling the acceleration sensor to monitor the acceleration change of the target falling object and obtaining the first target signal.

[0147] In practice, when the system initially determines that a falling object event may occur, the processor begins to perform integration on the first target signal a(t). The integration formula is: v(t)=∫a(t)dt+v0, where v(t) is the instantaneous velocity of the object at time t, and v0 is the initial velocity (in the case of free fall, the initial velocity is usually 0).

[0148] Based on the above embodiments, the processor can also compare the calculated instantaneous velocity v(t) with a preset velocity threshold v thresh Compare them. If v(t) ≥ v threshThis further confirms the falling object incident and incorporates speed information into the event analysis, providing support for subsequent sound source localization and object type identification. Specifically, to ensure the accuracy and reliability of the event judgment, the processor combines v(t) with the falling object location and sound signal intensity features obtained from the time difference localization algorithm to perform multi-dimensional verification of the falling object incident.

[0149] In the above embodiments, to ensure the accuracy of data analysis, in an optional implementation, this application sets up a system that can adjust its detection and response strategies by calculating the falling velocity v(t) in real time. For example, when the falling velocity is abnormally high, the system can automatically raise the alarm level to ensure that emergencies are handled promptly. The falling velocity information v(t) is not only used for real-time monitoring and event verification, but also for generating event reports. The report contains the velocity data of the falling object, helping managers assess the impact force and potential hazards of the falling object, providing a quantitative basis for safety management and risk prevention. The calculation of the falling velocity helps to further determine the degree of danger of the falling object, and is applicable to infrastructure such as bridges and tunnels that require assessment of the impact force of falling objects, providing data support for maintaining structural safety.

[0150] To facilitate the handling of fall incidents, in one optional embodiment, the above-mentioned device further includes:

[0151] The sixth determining unit is used to determine the type of the falling object as a heavy object when the target frequency is greater than or equal to the third threshold and less than or equal to the fourth threshold after verifying the first target signal based on the third target signal, and to determine the type of the falling object as a light object when the target frequency is greater than the fourth threshold and less than or equal to the fifth threshold.

[0152] In practice, the processor sets a third threshold F3 and a fourth threshold F4 to distinguish the frequency characteristics of heavy and light objects. These thresholds are set according to a preset acoustic model for object types, typically F3 = 20Hz and F4 = 10kHz, to accommodate the falling frequency range of different objects. The processor performs frequency analysis on the second target signal collected by the sound sensor, extracting the target frequency f. By comparing f with F3 and F4, the type of falling object is determined. If the target frequency f is greater than or equal to F3 and less than or equal to F4, the processor identifies the falling object as a heavy object; if f is greater than F4 and less than or equal to the fifth threshold F5 (e.g., F5 = 20kHz), the target object is classified as a light object. This process ensures that the system can intelligently identify the object type based on the sound characteristics of the falling object, providing crucial information for event handling.

[0153] The second calculation unit is used to determine the falling height of the target falling object based on the target location information, determine the falling time of the target falling object based on the falling height and the falling speed, and determine the falling position of the target falling object based on the target location information, the falling speed and the falling time.

[0154] Specifically, the processor is based on target location information (x s y s , z s Determine the falling height h of the target object. The falling height h can be calculated from the vertical coordinate difference in the target's position information, i.e., h = Hz. s Here, H represents the height of the pole or monitoring area. Using the free-fall motion formula and the known fall height h, the processor can deduce the fall time t. By calculating the fall time, the system can more accurately determine the entire process of an object's fall. The processor combines the fall velocity v(t), fall time t, and target position information to perform three-dimensional spatial positioning, determining the precise fall location of the target object. This process ensures that the system can provide detailed spatial information about the falling object event, supporting subsequent event analysis and processing.

[0155] The generation unit is used to generate alarm information based on the type of the falling object, the falling time, and the falling location.

[0156] In practice, the processor constructs alarm information based on information such as the type of falling object, the time of fall, and the location of fall, including key parameters such as the time of fall, object type, fall height, and precise fall location.

[0157] In the above embodiments, the system automatically records detailed information for each falling object incident, including alarm information, the type of falling object, the time and location of the fall, and generates an incident report. This data is used not only for immediate response but also for subsequent incident analysis, helping managers identify falling object risk trends and formulate effective risk prevention strategies. This type identification and detailed information recording function enables the system to provide comprehensive information after detecting a falling object, making it suitable for scenarios such as high-rise building management and urban security monitoring that require detailed accident reports, providing strong evidence for accident investigation and liability determination.

[0158] The aforementioned multi-source data fusion high-altitude falling object detection device includes a processor and a memory. The first control unit, second control unit, and third control unit, etc., are all stored as program units in the memory, and the processor executes the program units stored in the memory to achieve the corresponding functions. All of the above modules are located in the same processor; or, the above modules are located in different processors in any combination.

[0159] The processor contains a kernel, which retrieves the corresponding program units from memory. One or more kernels can be configured, and adjusting kernel parameters can improve the accuracy of diagnosing falling objects from heights.

[0160] The memory may include non-permanent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0161] This invention provides a computer-readable storage medium including a stored program, wherein, when the program is executed, it controls the device containing the computer-readable storage medium to perform the multi-source data fusion method for detecting falling objects from heights.

[0162] This application also provides a computer program product that, when executed on a data processing device, is adapted to perform a program that initializes the high-altitude falling object detection method steps involving at least the above-described multi-source data fusion.

[0163] It is obvious to those skilled in the art that the modules or steps of the present invention described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. They can be implemented using computer-executable program code, and thus can be stored in a storage device for execution by a computing device. In some cases, the steps shown or described can be performed in a different order than those described herein, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, the present invention is not limited to any particular combination of hardware and software.

[0164] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0165] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in one or more flowchart illustrations and / or one or more block diagrams.

[0166] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means that implement the functions specified in one or more flowcharts and / or one or more block diagrams.

[0167] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, such that the instructions, which execute on the computer or other programmable apparatus, provide steps for implementing the functions specified in one or more flowcharts and / or one or more block diagrams.

[0168] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0169] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0170] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0171] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0172] As can be seen from the above description, the embodiments of this application achieve the following technical effects:

[0173] 1) The multi-source data fusion system for detecting falling objects from heights in this application combines data from accelerometers and sound sensors, improving the accuracy and reliability of falling object detection. Accelerometers can quickly respond to falling object events, while sound sensors can locate the sound source. The combination of these two sensors not only allows for timely detection of falling objects but also accurately determines the type and location of the object, effectively avoiding misjudgments that may occur with a single sensor. Furthermore, verification through data fusion algorithms further improves detection accuracy, providing strong technical support for the safety management of falling objects from heights. In practical applications, this system can significantly improve public safety and reduce potential risks and losses caused by falling objects. The above embodiments solve the problem of low accuracy in object falling detection using a single type of sensor in existing technologies.

[0174] 2) The high-altitude falling object detection method based on multi-source data fusion of this application firstly controls the aforementioned accelerometer to monitor the acceleration changes of the target falling object, obtaining a first target signal, and determines whether a falling object event exists based on the first target signal; then, if the aforementioned accelerometer determines that a falling object event exists, it controls each of the aforementioned sound sensors to monitor the sound signal of the target falling object, obtaining multiple second target signals; finally, it controls the aforementioned processor to receive the aforementioned first target signal and the aforementioned second target signals, and performs data fusion based on the aforementioned first target signal and the aforementioned second target signals to obtain a third target signal, and verifies the aforementioned first target signal based on the third target signal. This application detects falling events by controlling the accelerometer and locates the falling object by controlling the sound sensor, and verifies the authenticity of the falling event by controlling the processor to perform data fusion on the monitoring signals of the accelerometer and the sound sensor, thus ensuring the accuracy of falling event detection and solving the problem of low accuracy in object falling detection using a single type of sensor in the prior art.

[0175] 3) The multi-source data fusion high-altitude falling object detection device of this application includes a first control unit that controls the aforementioned accelerometer to monitor the acceleration changes of a falling target object, obtaining a first target signal, and determining whether a falling object event exists based on the first target signal; a second control unit that, when the accelerometer determines that a falling object event exists, controls each of the aforementioned sound sensors to monitor the sound signal of the falling target object, obtaining multiple second target signals; and a third control unit that controls the aforementioned processor to receive the first target signal and the aforementioned second target signals, and performs data fusion based on the first target signal and the aforementioned second target signals to obtain a third target signal, and verifies the aforementioned first target signal based on the third target signal. This application detects falling events by controlling the accelerometer and locates falling objects by controlling the sound sensor, and verifies the authenticity of the falling event by controlling the processor to perform data fusion on the monitoring signals of the accelerometer and the sound sensor, thus ensuring the accuracy of falling event detection and solving the problem of low accuracy in object falling detection using a single type of sensor in the prior art.

[0176] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A high-altitude falling object detection system based on multi-source data fusion, characterized in that, include: An accelerometer is used to monitor changes in the acceleration of a falling object and to make a preliminary determination of whether a falling object event has occurred. A sound sensor, communicatively connected to the accelerometer, is used to monitor the sound signal generated by the falling object when the falling object event is detected, so as to locate the sound source and determine the type of the falling object. The processor is communicatively connected to the accelerometer and the sound sensor, and is used to receive sensor data and re-verify the falling object event through a data fusion algorithm to avoid misjudgment.

2. The system according to claim 1, characterized in that, The high-altitude falling object detection system also includes: An alarm module is communicatively connected to the processor. When the processor determines that the falling object event has occurred, it issues an alarm message to prompt the staff to handle the situation. The alarm message includes at least the time of the fall, the type of the falling object, and the location of the fall. The power supply module is electrically connected to the accelerometer, the sound sensor, the processor, and the alarm module, and is used to supply power to the accelerometer, the sound sensor, the processor, and the alarm module.

3. A method for detecting falling objects from heights using multi-source data fusion, characterized in that, The method for detecting falling objects from heights is used in the falling object detection system according to claim 1 or 2, and the method includes: The accelerometer is controlled to monitor the acceleration changes of the falling object, obtain a first target signal, and determine whether a falling object event exists based on the first target signal; When the accelerometer determines that the falling object event exists, it controls each of the sound sensors to monitor the sound signal of the falling target object and obtains multiple second target signals. The processor is controlled to receive the first target signal and the second target signal, and to perform data fusion based on the first target signal and the second target signal to obtain a third target signal, and to verify the first target signal based on the third target signal.

4. The method according to claim 3, characterized in that, Determining whether a falling object event exists based on the first target signal includes: Calculate the product of gravitational acceleration and a preset ratio to obtain the first threshold. If the acceleration of the falling object is greater than or equal to the first threshold, the falling object event is determined to exist.

5. The method according to claim 3, characterized in that, A third target signal is obtained by fusing data based on the first target signal and the second target signal, including: Based on the first target signal, the falling acceleration of the target falling object is determined; Based on the second target signal, determine the target sound pressure level of the sound generated by the falling object; Obtain a first preset weight and a second preset weight, wherein the first preset weight and the second preset weight are respectively weight parameters of the fall acceleration and the target sound pressure level; The third target signal is obtained by weighting the fall acceleration, the target sound pressure level, the first preset weight, and the second preset weight. If the third target signal is greater than or equal to the second threshold, it is determined that the falling object event exists.

6. The method according to claim 3, characterized in that, After controlling each of the aforementioned sound sensors to monitor the sound signal of the falling target object and obtaining multiple second target signals, the method further includes: Determine the time when the sound sensor corresponding to each second target signal receives the second target signal to obtain multiple target times; Based on the target time, the time difference between the reception of the second target signal by each of the sound sensors is determined to obtain the target duration; Based on the target duration, the distance difference between each sound sensor and the target falling object is determined to obtain the first target distance; The positional deviation between each of the sound sensors is obtained to get the second target distance. Based on the first target distance and the second target distance, the falling target object is located to obtain the target position information. The frequency of the sound produced by the falling target object is determined based on the second target signal to obtain the target frequency.

7. The method according to claim 6, characterized in that, After controlling the accelerometer to monitor the acceleration changes of the falling target object and obtain the first target signal, the method further includes: The falling speed of the target object is determined by integrating the first target signal.

8. The method according to claim 7, characterized in that, After verifying the first target signal based on the third target signal, the method further includes: If the target frequency is greater than or equal to the third threshold and less than or equal to the fourth threshold, the type of the target falling object is determined to be a heavy object; if the target frequency is greater than the fourth threshold and less than or equal to the fifth threshold, the type of the target falling object is determined to be a light object. The falling height of the target falling object is determined based on the target location information; the falling time of the target falling object is determined based on the falling height and the falling speed; and the falling position of the target falling object is determined based on the target location information, the falling speed, and the falling time. An alarm message is generated based on the type of the falling object, the time of the fall, and the location of the fall.

9. A high-altitude falling object detection device based on multi-source data fusion, characterized in that, The falling object detection device is used in the falling object detection system according to claim 1 or 2, and the device comprises: The first control unit is used to control the accelerometer to monitor the acceleration change of the target falling object, obtain the first target signal, and determine whether there is a falling object event based on the first target signal; The second control unit is used to control each of the sound sensors to monitor the sound signal of the target falling object when the acceleration sensor determines that the falling object event exists, so as to obtain multiple second target signals; The third control unit is used to control the processor to receive the first target signal and the second target signal, perform data fusion based on the first target signal and the second target signal to obtain a third target signal, and verify the first target signal based on the third target signal.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein, when the program is executed, it controls the device on which the computer-readable storage medium is located to perform the method of any one of claims 3 to 7.

Images