A personnel information monitoring method and system based on construction

Abstract

The present application relates to the field of construction management technology, and in particular to a personnel information monitoring method and system based on construction. The method includes: when performing identity authentication, controlling the personnel positioning card to collect the first voice command of the construction personnel, extracting the current voiceprint features in the first voice command; processing and analyzing the current voiceprint features to obtain a first analysis result, and determining whether the construction personnel can pass the identity authentication according to the first analysis result; when performing construction, controlling the personnel positioning card to regularly collect the second voice command of the construction personnel, extracting the current sound features in the second voice command; processing and analyzing the current sound features to obtain a second analysis result, and determining the current physical state of the construction personnel according to the second analysis result. By introducing voiceprint recognition technology, construction personnel can complete identity authentication without frequently taking off and putting on safety helmets, avoiding conflicts between safety equipment and construction equipment, and improving traffic efficiency and safety.

Description

Technical Field

[0001] The present application relates to the technical field of construction management, and in particular to a personnel information monitoring method and system based on building construction. Background Art

[0002] In today's rapidly developing construction industry, construction site gates are the core link of safety management, and their efficiency and safety are crucial to the smooth progress of engineering projects. With the advancement of technology, facial recognition technology has been introduced into construction site gate management, aiming to optimize workers' attendance and access processes and improve management efficiency. However, the application of this technology has encountered specific technical difficulties in practice, and the construction site management system has also exposed deficiencies in other aspects.

[0003] First, the application of facial recognition technology at construction site gates faces a unique challenge: the conflict between workers' safety equipment and the construction equipment they carry. In order to ensure the safety of construction site operations, workers must wear hard hats with good protective performance. However, when workers wear hard hats and hold construction equipment (such as sandblasters, grinders, etc.) through the facial recognition gate, the occlusion of the hard hats seriously affects the recognition effect of the facial recognition system. This causes workers to frequently take off their hard hats for identity verification and then put them back on, which not only increases the workers' operating burden, but also prolongs the passage time, affecting the passage efficiency of the gate, especially during the peak period of workers entering and leaving, and may also cause congestion at the gate.

[0004] On the other hand, the existing construction site management system has obvious deficiencies in the health monitoring of construction workers. The construction environment is complex and changeable. Workers are exposed to harsh conditions such as high temperature, high humidity, and high noise for a long time, facing a variety of health risks such as heat stroke and physical exhaustion. However, the current construction site management system lacks effective real-time monitoring methods to promptly detect and deal with these health problems. Especially in hot weather, the risk of heat stroke for workers increases significantly, but due to the lack of monitoring mechanisms, managers often find it difficult to grasp the physical condition of workers in a timely manner, thus missing the best time for intervention. Summary of the invention

[0005] The present application provides a personnel information monitoring method and system based on construction to solve the problem of conflict between workers' safety equipment and construction equipment in the process of optimizing gate management using face recognition technology in the existing construction industry.

[0006] In a first aspect, the present application provides a method for monitoring personnel information based on construction, which is applied to a personnel location card, comprising:

[0007] When performing identity authentication, controlling the personnel location card to collect a first voice command of the construction personnel, extracting a current voiceprint feature in the first voice command; processing and analyzing the current voiceprint feature to obtain a first analysis result, and determining whether the construction personnel can pass the identity authentication according to the first analysis result;

[0008] When performing construction, controlling the personnel location card to periodically collect the second voice command of the construction personnel and extracting the current sound features in the second voice command;

[0009] Processing and analyzing the current sound feature to obtain a second analysis result, and determining the current physical state of the construction worker according to the second analysis result; comparing the current physical state of the construction worker with a preset normal state;

[0010] If the current physical condition of the construction worker is lower than normal, the location information of the construction worker is obtained based on the personnel positioning card, and the construction worker is rescued according to the location information.

[0011] The above technical solution adopted in this application has the following advantages: by introducing voiceprint recognition technology, construction workers can complete identity verification without frequently taking off and putting on safety helmets, effectively avoiding conflicts between safety equipment and construction equipment, and improving traffic efficiency and safety. In addition, regular collection and analysis of the voice characteristics of construction workers can timely discover and warn of potential health problems, such as heat stroke and fatigue, providing important health monitoring data support for managers and ensuring the safety of construction workers.

[0012] Optionally, the processing and analyzing the current voiceprint feature to obtain a first analysis result, and determining whether the construction worker can pass identity authentication according to the first analysis result, includes:

[0013] Extracting the reserved voiceprint feature of the personnel location card, and matching the current voiceprint feature with the reserved voiceprint feature;

[0014] If the current voiceprint feature matches the reserved voiceprint feature, it is determined that the construction worker has passed the identity authentication;

[0015] If the current voiceprint feature does not match the reserved voiceprint feature, generating a simulated voiceprint feature based on the voiceprint model and the reserved voiceprint feature;

[0016] Matching the current voiceprint feature with the simulated voiceprint feature;

[0017] If the current voiceprint feature matches the simulated voiceprint feature, it is determined that the construction worker's identity verification has passed;

[0018] If the current voiceprint feature does not match the simulated voiceprint feature, it is determined that the construction worker's identity verification has failed.

[0019] The above technical solution adopted in this application has the following advantages: by introducing a simulated voiceprint generation and matching mechanism, it is able to tolerate natural changes or slight interference in voiceprint features to a certain extent, thereby improving the accuracy of identity verification. Moreover, even if the initial match fails, the verification will not be rejected immediately, but the identity will be further confirmed through simulated voiceprint generation and matching, which increases the complexity and security of the verification process. In addition, for construction workers whose initial match fails due to special reasons (such as slight changes in voice), more verification opportunities can be provided, avoiding unnecessary obstacles to passage and improving user experience. This method can flexibly respond to identity verification needs in different situations, and can give reasonable processing results whether it is fast passage under normal circumstances or complex verification under special circumstances.

[0020] Optionally, the personnel location card includes a plurality of microphone units, and the step of controlling the personnel location card to periodically collect the second voice command of the construction worker includes:

[0021] Controlling the personnel positioning card to send a voice collection signal to the construction personnel;

[0022] When the construction worker sends the voice collection signal, controlling the plurality of microphone units to simultaneously collect the voice signal sent by the construction worker;

[0023] Performing noise reduction processing on the speech signal to obtain a processed speech signal;

[0024] Performing feature extraction on the processed speech signal to obtain background noise and target speech;

[0025] Based on the speech separation technology, the target speech is extracted from the background noise to obtain a second voice instruction.

[0026] The above technical solution adopted in this application has the following advantages: through multi-microphone synchronous acquisition and noise reduction processing, the impact of environmental noise on voice quality is effectively reduced, and the clarity and recognizability of voice commands are improved. In addition, the target voice is extracted from the complex sound environment using voice separation technology to ensure the accuracy and reliability of health monitoring data. In addition, construction workers do not need to issue voice commands in a quiet environment, and can smoothly carry out health monitoring even on noisy construction sites, which improves the convenience and comfort of users.

[0027] Optionally, the processing and analyzing the current sound feature to obtain a second analysis result, and determining the current physical state of the construction worker according to the second analysis result includes:

[0028] Extracting body state features related to body state from the current sound features;

[0029] Based on the deep learning model, the body state characteristics are trained and analyzed to obtain body state data;

[0030] The current physical condition of the construction worker is determined according to the physical condition data.

[0031] The above technical solution adopted in this application has the following advantages: through the complex learning and analysis capabilities of the deep learning model, it is possible to more accurately extract information related to the physical state from the sound features, thereby improving the accuracy of the physical state assessment. In addition, the deep learning model has a strong generalization ability and can adapt to the individual differences and sound characteristics of different construction workers, making the assessment results more objective and reliable. In addition, it is possible to monitor the physical state of the construction workers in real time and issue early warnings in time when abnormal conditions are found, providing managers with timely intervention opportunities to ensure the life safety and health of the construction workers.

[0032] Optionally, acquiring the location information of the construction worker based on the personnel location card and rescuing the construction worker according to the location information includes:

[0033] Determine the location information of the construction personnel according to the personnel location card;

[0034] Obtain other location information of several other personnel location cards;

[0035] Determine other construction workers who are at the shortest distance from the construction worker according to the position information and the other positioning information;

[0036] The location information is sent to the other personnel positioning card, so that the other construction personnel move toward the construction personnel according to the location information and provide rescue.

[0037] The above technical solution adopted in this application has the following advantages: through accurate location information and fast algorithm matching, the nearest rescuer can be found quickly, the rescue response time can be shortened, and the rescue efficiency can be improved. In addition, the automation and real-time nature of the emergency rescue process enable site managers and construction workers to better respond to emergencies, reduce accident risks, and enhance the overall safety of the construction site. At the same time, in an emergency, the construction workers on the construction site can be automatically coordinated to promote teamwork and mutual assistance and jointly meet challenges. In addition, through accurate location tracking and rescuer allocation, the allocation of rescue resources can be optimized to ensure the efficiency and orderliness of rescue operations.

[0038] Optionally, the method further includes:

[0039] When a safety incident occurs, obtain some safety exit information;

[0040] Determine, based on the plurality of safety exit information and the location information, a nearest safety exit with the shortest distance from the location information;

[0041] According to the location information and the nearest safe exit, an optimal evacuation route is planned;

[0042] The optimal evacuation route is sent to the personnel positioning card so that the construction personnel can evacuate according to the optimal evacuation route.

[0043] The above technical solution adopted in this application has the following advantages: through intelligent path planning and real-time location tracking, it can provide the best evacuation route for construction workers, shorten the evacuation time, and improve the evacuation efficiency. In addition, it can ensure that construction workers can evacuate to a safe area quickly and orderly in an emergency, reducing casualties and losses.

[0044] Optionally, the personnel positioning card includes an acceleration sensor, and the method further includes:

[0045] According to the personnel positioning card, the current movement trajectory of the construction personnel is obtained in real time;

[0046] matching the current moving trajectory with the preset moving trajectory in real time, and when the current moving trajectory does not match the preset moving trajectory, controlling the acceleration sensor to perform acceleration detection on the current moving trajectory to obtain a current acceleration of the current moving trajectory;

[0047] The current acceleration is compared with the preset acceleration. If the current acceleration is greater than the preset acceleration, it is determined that there is an abnormality in the current acceleration, and the protective equipment of the construction personnel is controlled to perform emergency braking.

[0048] The above technical solution adopted in this application has the following advantages: through real-time monitoring of movement trajectory and acceleration detection, it is possible to timely detect and respond to abnormal movement behavior of construction personnel, effectively preventing the occurrence of safety accidents such as falls and collisions. In addition, the rapid response of the acceleration sensor and the instant triggering of the emergency braking function can take measures to protect the safety of construction personnel in the shortest time. In addition, the combination of protective equipment and intelligent monitoring system realizes the intelligent and automated control of protective equipment, and improves its efficiency and reliability in practical applications.

[0049] In a second aspect, the present application provides a personnel information monitoring system based on construction, comprising:

[0050] The voiceprint verification module is used to control the personnel positioning card to collect the first voice command of the construction personnel when performing identity authentication, extract the current voiceprint feature in the first voice command; process and analyze the current voiceprint feature to obtain a first analysis result, and determine whether the construction personnel can pass the identity authentication according to the first analysis result;

[0051] A sound feature analysis module, used to control the personnel location card to regularly collect the second voice command of the construction personnel during construction, and extract the current sound features in the second voice command;

[0052] A state comparison and evaluation module, used for processing and analyzing the current sound feature to obtain a second analysis result, determining the current physical state of the construction worker according to the second analysis result; and comparing the current physical state of the construction worker with a preset normal state;

[0053] The positioning and rescue module is used to obtain the location information of the construction worker based on the personnel positioning card if the current physical condition of the construction worker is lower than the normal state, and rescue the construction worker according to the location information.

[0054] Optionally, the voiceprint verification module is specifically used to:

[0055] Extracting the reserved voiceprint feature of the personnel location card, and matching the current voiceprint feature with the reserved voiceprint feature;

[0056] If the current voiceprint feature matches the reserved voiceprint feature, it is determined that the construction worker has passed the identity authentication;

[0057] If the current voiceprint feature does not match the reserved voiceprint feature, generating a simulated voiceprint feature based on the voiceprint model and the reserved voiceprint feature;

[0058] Matching the current voiceprint feature with the simulated voiceprint feature;

[0059] If the current voiceprint feature matches the simulated voiceprint feature, it is determined that the construction worker's identity verification has passed;

[0060] If the current voiceprint feature does not match the simulated voiceprint feature, it is determined that the construction worker's identity verification has failed.

[0061] Optionally, the sound feature analysis module is specifically used to:

[0062] Controlling the personnel positioning card to send a voice collection signal to the construction personnel;

[0063] When the construction worker sends the voice collection signal, controlling the plurality of microphone units to simultaneously collect the voice signal sent by the construction worker;

[0064] Performing noise reduction processing on the speech signal to obtain a processed speech signal;

[0065] Performing feature extraction on the processed speech signal to obtain background noise and target speech;

[0066] Based on the speech separation technology, the target speech is extracted from the background noise to obtain a second voice instruction.

[0067] Optionally, the state comparison and evaluation module is specifically used to:

[0068] Extracting body state features related to body state from the current sound features;

[0069] Based on the deep learning model, the body state characteristics are trained and analyzed to obtain body state data;

[0070] The current physical condition of the construction worker is determined according to the physical condition data.

[0071] Optionally, the positioning and rescue module is specifically used to:

[0072] Determine the location information of the construction personnel according to the personnel location card;

[0073] Obtain other location information of several other personnel location cards;

[0074] Determine other construction workers who are at the shortest distance from the construction worker according to the position information and the other positioning information;

[0075] The location information is sent to the other personnel positioning card, so that the other construction personnel move toward the construction personnel according to the location information and provide rescue.

[0076] Optionally, the personnel information monitoring system further includes an accident handling module, which is used to:

[0077] When a safety incident occurs, obtain some safety exit information;

[0078] Determine, based on the plurality of safety exit information and the location information, a nearest safety exit with the shortest distance from the location information;

[0079] According to the location information and the nearest safe exit, an optimal evacuation route is planned;

[0080] The optimal evacuation route is sent to the personnel positioning card so that the construction personnel can evacuate according to the optimal evacuation route.

[0081] Optionally, the personnel information monitoring system further includes a protection module, which is used to:

[0082] According to the personnel positioning card, the current movement trajectory of the construction personnel is obtained in real time;

[0083] matching the current moving trajectory with the preset moving trajectory in real time, and when the current moving trajectory does not match the preset moving trajectory, controlling the acceleration sensor to perform acceleration detection on the current moving trajectory to obtain a current acceleration of the current moving trajectory;

[0084] The current acceleration is compared with the preset acceleration. If the current acceleration is greater than the preset acceleration, it is determined that the current acceleration is abnormal, and the protective equipment of the construction personnel is controlled to perform emergency braking.

[0085] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0086] Figure 1 A schematic diagram of an application scenario provided for an embodiment of the present application;

[0087] Figure 2 A flowchart of a method for monitoring personnel information based on construction provided in an embodiment of the present application;

[0088] Figure 3 A schematic structural diagram of a personnel information monitoring system based on construction provided in one embodiment of the present application. DETAILED DESCRIPTION

[0089] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0090] In addition, the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article, unless otherwise specified, generally means that the associated objects before and after are in an "or" relationship.

[0091] The embodiments of the present application are further described in detail below in conjunction with the drawings in the specification.

[0092] In today's rapidly developing construction industry, construction site gates are the core link of safety management, and their efficiency and safety are crucial to the smooth progress of engineering projects. With the advancement of technology, facial recognition technology has been introduced into construction site gate management, aiming to optimize workers' attendance and access processes and improve management efficiency. However, the application of this technology has encountered specific technical difficulties in practice, and the construction site management system has also exposed deficiencies in other aspects.

[0093] First, the application of facial recognition technology at construction site gates faces a unique challenge: the conflict between workers' safety equipment and the construction equipment they carry. In order to ensure the safety of construction site operations, workers must wear hard hats with good protective performance. However, when workers wear hard hats and hold construction equipment (such as sandblasters, grinders, etc.) through the facial recognition gate, the occlusion of the hard hats seriously affects the recognition effect of the facial recognition system. This causes workers to frequently take off their hard hats for identity verification and then put them back on, which not only increases the workers' operating burden, but also prolongs the passage time, affecting the passage efficiency of the gate, especially during the peak period of workers entering and leaving, and may also cause congestion at the gate.

[0094] On the other hand, the existing construction site management system has obvious deficiencies in the health monitoring of construction workers. The construction environment is complex and changeable. Workers are exposed to harsh conditions such as high temperature, high humidity, and high noise for a long time, facing a variety of health risks such as heat stroke and physical exhaustion. However, the current construction site management system lacks effective real-time monitoring methods to promptly detect and deal with these health problems. Especially in hot weather, the risk of heat stroke for workers increases significantly, but due to the lack of monitoring mechanisms, managers often find it difficult to grasp the physical condition of workers in a timely manner, thus missing the best time for intervention.

[0095] Based on this, the present application provides a personnel information monitoring method and system based on construction. By introducing voiceprint recognition technology, construction workers can complete identity verification without frequently taking off and putting on safety helmets, effectively avoiding conflicts between safety equipment and construction equipment, and improving traffic efficiency and safety. In addition, regular collection and analysis of the voice characteristics of construction workers can timely discover and warn of potential health problems, such as heat stroke, fatigue, etc., providing important health monitoring data support for managers and ensuring the safety of construction workers.

[0096] Figure 1 A schematic diagram of an application scenario provided by the present application is used to manage the construction site during the construction site management process. Specifically, a server is set in a personnel information monitoring system based on construction, and the server communicates with a personnel positioning card to jointly implement a personnel information monitoring method based on construction of the present application, and the first voice command of the construction personnel is collected through the personnel positioning card to achieve the purpose of voiceprint verification of identity, and the second voice command of the construction personnel is collected through the personnel positioning card to monitor the physical state of the construction personnel, and the location information of the construction personnel is obtained through the personnel positioning card, so as to rescue the construction personnel according to the location information. By introducing voiceprint recognition technology, construction personnel can complete identity verification without frequently taking off and wearing safety helmets, effectively avoiding the conflict between safety equipment and construction equipment, and improving traffic efficiency and safety. In addition, the voice characteristics of construction personnel are regularly collected and analyzed, which can timely discover and warn potential health problems, such as heat stroke, fatigue, etc., and provide important health monitoring data support for management personnel, ensuring the life safety of construction personnel. The specific implementation method can refer to the following embodiments.

[0097] Figure 2 This is a flowchart of a method for monitoring personnel information based on construction provided in an embodiment of the present application. The method of this embodiment can be applied to the server in the above scenario. Figure 2 As shown, the method includes:

[0098] S201. When performing identity authentication, the control personnel positioning card collects the first voice command of the construction personnel and extracts the current voiceprint feature in the first voice command; the current voiceprint feature is processed and analyzed to obtain a first analysis result, and whether the construction personnel can pass the identity authentication is determined according to the first analysis result.

[0099] Specifically, when construction workers approach the construction site gate, the identity verification process of the facial recognition gate is triggered.

[0100] At this time, the construction worker issues a voice command (such as "Authentication, Zhang San") through the personnel positioning card (built-in microphone and processor) worn on his body as the first voice command. The voice command is collected through the personnel positioning card, and the voiceprint features are extracted, and then the voiceprint features are sent to the back-end server for comparison and analysis. The back-end server compares the voiceprint features with the pre-stored voiceprint database, and after confirming the identity of the construction worker, allows him to pass through the gate and records the passing time.

[0101] S202: When performing construction work, the control personnel positioning card periodically collects the second voice instructions of the construction personnel and extracts the current sound features in the second voice instructions.

[0102] Specifically, during the construction process, the personnel location card automatically wakes up at a preset time interval (such as every half an hour), prompting the construction personnel to issue a short voice as the second voice command (such as "status report"). After the construction personnel issue the command, the personnel location card collects the second voice command and extracts the sound features therein, including volume, pitch, speaking speed, etc. These features reflect the current physical state of the construction personnel (such as fatigue level, emotional state, etc.).

[0103] S203, processing and analyzing the current sound feature to obtain a second analysis result, determining the current physical state of the construction worker according to the second analysis result; and comparing the current physical state of the construction worker with a preset normal state.

[0104] Specifically, the sound features are processed and analyzed using a machine learning algorithm to generate a second analysis result.

[0105] Next, the second analysis result is compared with a preset normal state threshold value to assess whether the physical state of the construction worker is within a normal range.

[0106] S204: If the current physical condition of the construction worker is lower than the normal condition, the location information of the construction worker is obtained based on the personnel positioning card, and the construction worker is rescued according to the location information.

[0107] Specifically, if a construction worker's physical condition is found to be below the normal threshold (such as weak voice and slow speech, which may indicate fatigue or heat stroke), the personnel location card will immediately trigger an early warning. Based on the construction worker's location information obtained in real time by the personnel location card, the management can quickly locate the worker and dispatch a rescue team to the scene for assistance.

[0108] Through the method provided in this embodiment, by introducing voiceprint recognition technology, construction workers can complete identity verification without frequently taking off and putting on safety helmets, effectively avoiding conflicts between safety equipment and construction equipment, and improving traffic efficiency and safety. In addition, regular collection and analysis of the voice characteristics of construction workers can timely discover and warn of potential health problems, such as heat stroke and fatigue, providing important health monitoring data support for managers and ensuring the safety of construction workers.

[0109] In some embodiments, the reserved voiceprint features of the personnel location card are extracted, and the current voiceprint features are matched with the reserved voiceprint features; if the current voiceprint features match the reserved voiceprint features, it is determined that the construction personnel has passed the identity authentication; if the current voiceprint features do not match the reserved voiceprint features, then based on the voiceprint model, a simulated voiceprint feature is generated according to the reserved voiceprint features; the current voiceprint features are matched with the simulated voiceprint features; if the current voiceprint features match the simulated voiceprint features, it is determined that the construction personnel has passed the identity authentication; if the current voiceprint features do not match the simulated voiceprint features, it is determined that the construction personnel has failed the identity authentication.

[0110] Specifically, after the construction worker issues a voice command as the first voice command, the voiceprint feature reserved by the personnel positioning card worn by the construction worker is first extracted, and the current voiceprint feature is directly matched with the reserved voiceprint feature. If the two are matched successfully, it is directly determined that the construction worker has passed the identity verification and is allowed to pass.

[0111] If the initial match fails, verification will not be rejected immediately, but will enter a more complex verification process.

[0112] Specifically, based on the voiceprint model (which may use deep learning, machine learning and other technologies to simulate the changes in voiceprint features), simulated voiceprint features are generated according to the reserved voiceprint features. These simulated features may take into account environmental factors (such as noise), physical factors (voice changes caused by throat diseases), time changes (such as changes in voice with age), etc.

[0113] Next, the current voiceprint feature is matched with these simulated voiceprint features to try to find the best matching simulated voiceprint. If the current voiceprint feature successfully matches a simulated voiceprint feature, it is considered that the construction worker has passed the identity verification and is allowed to pass.

[0114] If after all the above steps, the current voiceprint feature still cannot match the reserved voiceprint feature or any simulated voiceprint feature, it is determined that the construction worker's identity authentication has failed, and his / her passage is denied, and the relevant information is recorded for subsequent review.

[0115] Through the solution provided by this embodiment, by introducing a simulated voiceprint generation and matching mechanism, it is possible to tolerate natural changes or slight interference in voiceprint features to a certain extent, thereby improving the accuracy of identity verification. Moreover, even if the initial match fails, the verification will not be rejected immediately, but the identity will be further confirmed through simulated voiceprint generation and matching, which increases the complexity and security of the verification process. In addition, for construction workers whose initial matching fails due to special reasons (such as slight changes in voice), more verification opportunities can be provided, avoiding unnecessary obstacles to passage and improving user experience. This method can flexibly respond to identity verification needs in different situations, and can give reasonable processing results whether it is fast passage under normal circumstances or complex verification under special circumstances.

[0116] In some embodiments, the personnel positioning card is controlled to send a voice collection signal to the construction personnel; when the construction personnel send the voice collection signal, several microphone units are controlled to simultaneously collect the voice signal sent by the construction personnel; the voice signal is subjected to noise reduction processing to obtain a processed voice signal; feature extraction is performed on the processed voice signal to obtain background noise and target voice; based on voice separation technology, the target voice is extracted from the background noise to obtain a second voice instruction.

[0117] The personnel locator card contains several microphone units.

[0118] Specifically, when the preset time interval arrives, the personnel positioning card sends a voice collection signal to the construction personnel through the built-in audio playback device, prompting them to issue a second voice command (such as "status report").

[0119] When the construction worker responds to the voice collection signal and speaks, the personnel positioning card controls the multiple microphone units on it to simultaneously collect the voice signals emitted by the construction worker. The multi-microphone configuration helps capture more comprehensive sound information, including the directionality and spatial characteristics of the sound.

[0120] Next, the collected speech signal is subjected to noise reduction processing to reduce the impact of environmental noise on speech quality. Noise reduction processing may include using digital signal processing algorithms (such as adaptive filtering, spectral subtraction, etc.) to identify and suppress background noise.

[0121] Next, feature extraction is performed on the denoised speech signal to distinguish the background noise from the target speech. Feature extraction may involve techniques such as spectrum analysis and time-frequency analysis of audio signals.

[0122] Next, the target speech is separated from the background noise based on speech separation techniques (such as blind source separation, computational auditory scene analysis, etc.). This step ensures the accuracy of subsequent analysis, especially when the construction environment is noisy.

[0123] Finally, after the above processing, a clear and pure second voice instruction is obtained, which is then used for analysis and evaluation of health status.

[0124] Through the solution provided in this embodiment, the impact of environmental noise on voice quality is effectively reduced through multi-microphone synchronous acquisition and noise reduction processing, and the clarity and recognizability of voice commands are improved. In addition, the target voice is extracted from the complex sound environment using voice separation technology to ensure the accuracy and reliability of health monitoring data. In addition, construction workers do not need to issue voice commands in a quiet environment, and can smoothly perform health monitoring even on noisy construction sites, which improves user convenience and comfort.

[0125] In some embodiments, body state features related to body state are extracted from current sound features; based on a deep learning model, the body state features are trained and analyzed to obtain body state data; and based on the body state data, the current body state of the construction worker is determined.

[0126] Specifically, after extracting the current sound features from the second voice command, further screen out features closely related to the physical state, such as volume, pitch, speech speed, voice tremor, etc. These features can reflect the fatigue level, emotional state and possible health problems of the construction workers.

[0127] Next, a deep learning model is pre-trained, which uses a large amount of speech data containing different physical states (such as health, fatigue, tension, illness, etc.) as a training set. By learning the feature representation of this data, the model can establish a complex mapping relationship between sound features and physical states.

[0128] Next, the extracted physical state features are input into the trained deep learning model, which performs complex nonlinear transformations and combinations on these features and finally outputs physical state data, which may include quantitative indicators such as fatigue index, emotional score, and health risk.

[0129] Next, based on the physical state data output by the deep learning model, a series of thresholds or classification rules can be set to determine the current physical state of the construction worker. For example, if the fatigue index exceeds a certain threshold, the construction worker is judged to be in a fatigued state; if the emotional score is low, it may mean that the construction worker is depressed or nervous.

[0130] Through the scheme provided by this embodiment, through the complex learning and analysis capabilities of the deep learning model, it is possible to more accurately extract information related to the physical state from the sound features, thereby improving the accuracy of the physical state assessment. In addition, the deep learning model has a strong generalization ability and can adapt to the individual differences and sound characteristics of different construction workers, making the assessment results more objective and reliable. In addition, it is possible to monitor the physical state of the construction workers in real time, and issue early warnings in time when abnormal conditions are found, providing managers with timely intervention opportunities to ensure the life safety and health of the construction workers.

[0131] In some embodiments, the location information of the construction workers is determined based on the personnel location card; other location information of several other personnel location cards is obtained; other construction workers who are the closest to the construction workers are determined based on the location information and other location information; the location information is sent to the other personnel location cards so that other construction workers can move towards the construction workers based on the location information and provide rescue.

[0132] Specifically, the precise location information of construction workers is obtained in real time through the built-in GPS or other positioning technologies (such as Wi-Fi positioning, Bluetooth beacons, etc.) of the personnel positioning card. At the same time, the personnel positioning card information of other construction workers on the construction site is collected to obtain their current location. This information constitutes a dynamic distribution map of construction site personnel.

[0133] Then, based on the location information of the construction worker and the location information of other construction workers, other construction workers with the shortest distance to the construction worker are quickly identified as potential rescuers by calculating the distance or using a geospatial algorithm.

[0134] The precise location of the construction worker is then sent to the personnel location card of the nearest rescuer, which may be achieved through wireless communication technology (such as Bluetooth, Wi-Fi Direct or mobile network).

[0135] Finally, the rescuers who received the location information quickly moved toward the construction workers according to the instructions, and carried out necessary rescue actions after arriving at the scene, such as providing medical assistance and evacuating dangerous areas.

[0136] Through the solution provided by this embodiment, through accurate location information and fast algorithm matching, the nearest rescuer can be quickly found, the rescue response time can be shortened, and the rescue efficiency can be improved. In addition, the automation and real-time nature of the emergency rescue process enable site managers and construction workers to better respond to emergencies, reduce accident risks, and enhance the overall safety of the site. At the same time, in an emergency, the construction workers on the site can be automatically coordinated to promote teamwork and mutual assistance and jointly respond to challenges. In addition, through accurate location tracking and rescuer allocation, the allocation of rescue resources can be optimized to ensure that the rescue operation is carried out efficiently and in an orderly manner.

[0137] In some embodiments, when a safety accident occurs, several safety exit information is obtained; based on the several safety exit information and location information, the nearest safety exit with the shortest distance to the location information is determined; based on the location information and the nearest safety exit, the optimal evacuation route is planned; and the optimal evacuation route is sent to the personnel positioning card so that the construction personnel can evacuate according to the optimal evacuation route.

[0138] Specifically, when a safety accident (such as fire, collapse, etc.) is detected, the emergency response mechanism is immediately activated.

[0139] First, obtain the location information of all safety exits on the construction site, including emergency exits, stairwells, escape routes, etc. In an emergency, this information will be quickly retrieved.

[0140] Next, based on the current location information of the construction workers and the location information of all safety exits, the safety exit closest to the construction workers is determined by calculating the distance or using a geospatial algorithm.

[0141] Next, considering factors such as obstacles on the construction site, crowd density, and safety exit capacity, a path planning algorithm (such as A* algorithm, Dijkstra algorithm, etc.) is used to plan an optimal evacuation route for the construction workers. The route should be as short as possible, safe, and not prone to congestion.

[0142] Then, the optimal evacuation route is sent to the construction worker's personnel location card, which may be achieved through wireless communication technology (such as Bluetooth, Wi-Fi direct or mobile network). At the same time, a sound or vibration prompt is issued to the construction worker to attract his attention.

[0143] Finally, after receiving the evacuation order, the construction workers quickly evacuated to the nearest safe exit according to the indicated optimal evacuation route, monitored the evacuation progress in real time, and provided feedback to the management staff.

[0144] Through the solution provided in this embodiment, through intelligent path planning and real-time location tracking, it is possible to provide construction workers with the best evacuation route, shorten evacuation time, and improve evacuation efficiency. In addition, it is possible to ensure that construction workers can evacuate to a safe area quickly and orderly in an emergency, reducing casualties and losses.

[0145] In some embodiments, the current movement trajectory of the construction workers is obtained in real time based on the personnel positioning card; the current movement trajectory is matched with the preset movement trajectory in real time. When the current movement trajectory does not match the preset movement trajectory, the acceleration sensor is controlled to perform acceleration detection on the current movement trajectory to obtain the current acceleration of the current movement trajectory; the current acceleration is compared with the preset acceleration. If the current acceleration is greater than the preset acceleration, it is determined that there is an abnormality in the current acceleration, and the construction workers' protective equipment is controlled to perform emergency braking.

[0146] The personnel positioning card contains an acceleration sensor.

[0147] Specifically, the current movement trajectory of the construction workers can be obtained in real time through the built-in GPS or other positioning technologies in the personnel positioning card, as well as possible inertial navigation systems (such as acceleration sensors, gyroscopes, etc.).

[0148] Next, the current moving trajectory is matched with the preset moving trajectory (such as normal construction path, safe evacuation route, etc.) in real time. The preset trajectory can be set in advance according to the construction site layout and work plan. When it is found that the current moving trajectory does not match the preset trajectory, it is considered that there may be an abnormal situation and further detection is required.

[0149] At this time, the acceleration sensor is triggered to detect the acceleration of the current moving trajectory and obtain the current acceleration information of the construction personnel. The acceleration sensor can sensitively capture the changes in the speed of personnel movement and provide important data for abnormal detection.

[0150] Next, the current acceleration is compared with the preset acceleration threshold. The preset acceleration can be set according to the actual situation of the construction site and the safety standards of personnel to reflect the speed range of personnel movement under normal circumstances. If the current acceleration is greater than the preset acceleration threshold, it is determined that the current acceleration is abnormal, which may mean that the construction personnel are performing dangerous or uncontrolled movements.

[0151] Once the current acceleration is determined to be abnormal, the protective equipment of the construction workers is immediately controlled for emergency braking. The protective equipment may include safety belts, fall prevention devices, emergency brake shoes, etc., depending on the site environment and work requirements. The emergency braking function sends instructions to the protective equipment through wireless communication technology (such as Bluetooth, Wi-Fi Direct or mobile network), triggering its braking mechanism to quickly fix the construction workers in a safe position or slow down their movement speed.

[0152] Through the solution provided by this embodiment, by real-time monitoring of movement trajectory and acceleration detection, abnormal movement behavior of construction personnel can be discovered and responded to in a timely manner, effectively preventing the occurrence of safety accidents such as falls and collisions. In addition, the rapid response of the acceleration sensor and the instant triggering of the emergency braking function can take measures to protect the safety of construction personnel in the shortest time. In addition, the combination of protective equipment and intelligent monitoring system realizes the intelligent and automated control of protective equipment, and improves its efficiency and reliability in practical applications.

[0153] Figure 3 A schematic diagram of a personnel information monitoring system based on construction provided in an embodiment of the present application is shown in FIG. Figure 3As shown, the personnel information monitoring system 300 of this embodiment includes: a voiceprint verification module 301, a sound feature analysis module 302, a state comparison and evaluation module 303, and a positioning and rescue module 304.

[0154] The voiceprint verification module 301 is used to control the personnel positioning card to collect the first voice command of the construction personnel when performing identity authentication, and extract the current voiceprint feature in the first voice command; process and analyze the current voiceprint feature to obtain a first analysis result, and determine whether the construction personnel can pass the identity authentication according to the first analysis result;

[0155] The sound feature analysis module 302 is used to control the personnel location card to regularly collect the second voice command of the construction personnel during construction, and extract the current sound feature in the second voice command;

[0156] The state comparison and evaluation module 303 is used to process and analyze the current sound feature to obtain a second analysis result, determine the current physical state of the construction worker according to the second analysis result; and compare the current physical state of the construction worker with a preset normal state;

[0157] The positioning and rescue module 304 is used to obtain the location information of the construction worker based on the personnel positioning card if the current physical condition of the construction worker is lower than the normal state, and rescue the construction worker according to the location information.

[0158] Optionally, the voiceprint verification module 301 is specifically used to:

[0159] Extracting the reserved voiceprint feature of the personnel location card, and matching the current voiceprint feature with the reserved voiceprint feature;

[0160] If the current voiceprint feature matches the reserved voiceprint feature, it is determined that the construction worker has passed the identity authentication;

[0161] If the current voiceprint feature does not match the reserved voiceprint feature, generating a simulated voiceprint feature based on the voiceprint model and the reserved voiceprint feature;

[0162] Matching the current voiceprint feature with the simulated voiceprint feature;

[0163] If the current voiceprint feature matches the simulated voiceprint feature, it is determined that the construction worker's identity verification has passed;

[0164] If the current voiceprint feature does not match the simulated voiceprint feature, it is determined that the construction worker's identity verification has failed.

[0165] Optionally, the sound feature analysis module 302 is specifically used to:

[0166] Controlling the personnel positioning card to send a voice collection signal to the construction personnel;

[0167] When the construction worker sends the voice collection signal, controlling the plurality of microphone units to simultaneously collect the voice signal sent by the construction worker;

[0168] Performing noise reduction processing on the speech signal to obtain a processed speech signal;

[0169] Performing feature extraction on the processed speech signal to obtain background noise and target speech;

[0170] Based on the speech separation technology, the target speech is extracted from the background noise to obtain a second voice instruction.

[0171] Optionally, the state comparison and evaluation module 303 is specifically used for:

[0172] Extracting body state features related to body state from the current sound features;

[0173] Based on the deep learning model, the body state characteristics are trained and analyzed to obtain body state data;

[0174] The current physical condition of the construction worker is determined according to the physical condition data.

[0175] Optionally, the positioning and rescue module 304 is specifically used to:

[0176] Determine the location information of the construction personnel according to the personnel location card;

[0177] Obtain other location information of several other personnel location cards;

[0178] Determine other construction workers who are at the shortest distance from the construction worker according to the position information and the other positioning information;

[0179] The location information is sent to the other personnel positioning card, so that the other construction personnel move toward the construction personnel according to the location information and provide rescue.

[0180] Optionally, the personnel information monitoring system 300 further includes an accident handling module 305, which is used to:

[0181] When a safety incident occurs, obtain some safety exit information;

[0182] Determine, based on the plurality of safety exit information and the location information, a nearest safety exit with the shortest distance from the location information;

[0183] According to the location information and the nearest safe exit, an optimal evacuation route is planned;

[0184] The optimal evacuation route is sent to the personnel positioning card so that the construction personnel can evacuate according to the optimal evacuation route.

[0185] Optionally, the personnel information monitoring system 300 further includes a protection module 306, which is used to:

[0186] According to the personnel positioning card, the current movement trajectory of the construction personnel is obtained in real time;

[0187] matching the current moving trajectory with the preset moving trajectory in real time, and when the current moving trajectory does not match the preset moving trajectory, controlling the acceleration sensor to perform acceleration detection on the current moving trajectory to obtain a current acceleration of the current moving trajectory;

[0188] The current acceleration is compared with the preset acceleration. If the current acceleration is greater than the preset acceleration, it is determined that there is an abnormality in the current acceleration, and the protective equipment of the construction personnel is controlled to perform emergency braking.

[0189] The system of this embodiment can be used to execute the method of any of the above embodiments. The implementation principles and technical effects are similar and will not be described in detail here.

Claims

1. A personnel information monitoring method based on construction, characterized in that: Applied to personnel location cards, including: When performing identity authentication, controlling the personnel location card to collect a first voice command of the construction personnel, extracting a current voiceprint feature in the first voice command; processing and analyzing the current voiceprint feature to obtain a first analysis result, and determining whether the construction personnel can pass the identity authentication according to the first analysis result; When performing construction, controlling the personnel location card to periodically collect the second voice command of the construction personnel and extracting the current sound features in the second voice command; Processing and analyzing the current sound feature to obtain a second analysis result, and determining the current physical state of the construction worker according to the second analysis result; comparing the current physical state of the construction worker with a preset normal state; If the current physical condition of the construction worker is lower than normal, obtaining the location information of the construction worker based on the personnel positioning card, and rescuing the construction worker according to the location information; The processing and analyzing of the current voiceprint feature to obtain a first analysis result, and determining whether the construction personnel can pass identity authentication according to the first analysis result, includes: Extracting the reserved voiceprint feature of the personnel location card, and matching the current voiceprint feature with the reserved voiceprint feature; If the current voiceprint feature matches the reserved voiceprint feature, it is determined that the construction worker has passed the identity authentication; If the current voiceprint feature does not match the reserved voiceprint feature, generating a simulated voiceprint feature based on the voiceprint model and the reserved voiceprint feature; Matching the current voiceprint feature with the simulated voiceprint feature; If the current voiceprint feature matches the simulated voiceprint feature, it is determined that the construction worker's identity verification has passed; If the current voiceprint feature does not match the simulated voiceprint feature, it is determined that the construction worker's identity verification has failed.

2. The method according to claim 1, characterized in that The personnel location card includes a plurality of microphone units, and the controlling the personnel location card to periodically collect the second voice command of the construction personnel includes: Controlling the personnel positioning card to send a voice collection signal to the construction personnel; When the construction worker sends the voice collection signal, controlling the plurality of microphone units to simultaneously collect the voice signal sent by the construction worker; Performing noise reduction processing on the speech signal to obtain a processed speech signal; Performing feature extraction on the processed speech signal to obtain background noise and target speech; Based on the speech separation technology, the target speech is extracted from the background noise to obtain a second voice instruction.

3. The method according to claim 2, characterized in that The processing and analyzing the current sound feature to obtain a second analysis result, and determining the current physical state of the construction worker according to the second analysis result includes: Extracting body state features related to body state from the current sound features; Based on the deep learning model, the body state characteristics are trained and analyzed to obtain body state data; The current physical condition of the construction worker is determined according to the physical condition data.

4. The method according to claim 3, characterized in that The obtaining the location information of the construction worker based on the personnel location card and rescuing the construction worker according to the location information includes: Determine the location information of the construction personnel according to the personnel location card; Obtain other location information of several other personnel location cards; Determine other construction workers who are at the shortest distance from the construction worker according to the position information and the other positioning information; The location information is sent to the positioning card of the other personnel, so that the other construction personnel can move toward the construction personnel according to the location information and provide rescue.

5. The method according to claim 4, characterized in that The method further comprises: When a safety incident occurs, obtain some safety exit information; Determine, based on the plurality of safety exit information and the location information, a nearest safety exit with the shortest distance from the location information; According to the location information and the nearest safe exit, an optimal evacuation route is planned; The optimal evacuation route is sent to the personnel positioning card so that the construction personnel can evacuate according to the optimal evacuation route.

6. The method according to claim 5, characterized in that The personnel positioning card includes an acceleration sensor, and the method further includes: According to the personnel positioning card, the current movement trajectory of the construction personnel is obtained in real time; matching the current moving trajectory with the preset moving trajectory in real time, and when the current moving trajectory does not match the preset moving trajectory, controlling the acceleration sensor to perform acceleration detection on the current moving trajectory to obtain a current acceleration of the current moving trajectory; The current acceleration is compared with the preset acceleration. If the current acceleration is greater than the preset acceleration, it is determined that there is an abnormality in the current acceleration, and the protective equipment of the construction personnel is controlled to perform emergency braking.

7. A personnel information monitoring system based on construction, characterized in that: include: A voiceprint verification module, used to control the personnel positioning card to collect the first voice command of the construction personnel when performing identity authentication, and extract the current voiceprint feature in the first voice command; Processing and analyzing the current voiceprint feature to obtain a first analysis result, and determining whether the construction worker can pass identity authentication according to the first analysis result; A sound feature analysis module, used to control the personnel location card to regularly collect the second voice command of the construction personnel during construction, and extract the current sound features in the second voice command; A state comparison and evaluation module, used for processing and analyzing the current sound feature to obtain a second analysis result, and determining the current physical state of the construction worker according to the second analysis result; Comparing the current physical state of the construction worker with a preset normal state; A positioning and rescue module, used for obtaining the location information of the construction worker based on the personnel positioning card if the current physical condition of the construction worker is lower than the normal state, and rescuing the construction worker according to the location information; The voiceprint verification module is further used to extract the reserved voiceprint feature of the personnel location card and match the current voiceprint feature with the reserved voiceprint feature; If the current voiceprint feature matches the reserved voiceprint feature, it is determined that the construction worker has passed the identity authentication; If the current voiceprint feature does not match the reserved voiceprint feature, generating a simulated voiceprint feature based on the voiceprint model and the reserved voiceprint feature; Matching the current voiceprint feature with the simulated voiceprint feature; If the current voiceprint feature matches the simulated voiceprint feature, it is determined that the construction worker's identity verification has passed; If the current voiceprint feature does not match the simulated voiceprint feature, it is determined that the construction worker's identity verification has failed.

Images