Construction personnel intelligent management method and system based on unmanned aerial vehicle

Abstract

The invention relates to an intelligent construction personnel management method and system based on an unmanned aerial vehicle, and relates to the technical field of unmanned aerial vehicles, and the method comprises the steps: carrying out the flight path planning of the unmanned aerial vehicle in a construction area, and obtaining a flight path; the unmanned aerial vehicle is controlled to fly in the construction area according to the flight path, a monitoring image set and a monitoring audio set are acquired, work type identification is carried out, and work type information distribution is obtained; performing risk identification according to the work type information distribution to obtain a risk identification result; and integrating the work type information distribution and the risk identification result to obtain a construction personnel monitoring result, and performing visual display. The technical problems that current unmanned aerial vehicle construction management is not intelligent, and construction safety risk management efficiency and quality are affected are effectively solved.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicles (UAVs), and in particular to an intelligent management method and system for construction personnel based on UAVs. Background Art

[0002] With the rapid development of the construction industry, large-scale, high-level housing construction projects are increasing in number. Construction sites are becoming increasingly complex, involving multiple construction areas and the collaborative work of various types of work. In this dynamic construction environment, how to efficiently manage construction personnel and ensure construction safety and progress has become a major challenge in construction management. Traditional construction personnel management relies primarily on manual inspections and video surveillance. However, these methods have certain limitations in practical application, such as low inspection efficiency, limited coverage, single data collection methods, incomplete information acquisition, and a lack of intelligent dynamic management and risk warning methods.

[0003] Most of the current drone applications remain at the level of simple image acquisition and video monitoring, lacking in-depth analysis. In a dynamic construction environment, drones are unable to optimize path planning and avoid interfering with the construction process. There are technical problems such as the unintelligent management of construction through drones, which affects the effectiveness of drone use and the efficiency and quality of construction safety risk management. Summary of the Invention

[0004] The present invention aims to solve the technical problems in the prior art that construction management through drones is not intelligent, which affects the use effect of drones and the efficiency and quality of construction safety risk management. It provides a drone-based intelligent management method and system for construction personnel to solve the problem.

[0005] The technical solution of the present invention to solve the above technical problems is as follows:

[0006] In a first aspect, the present invention provides a method for intelligent management of construction personnel based on drones, comprising: planning a flight path of the drone within a construction area to obtain a flight path;

[0007] Controlling the drone to fly within the construction area according to the flight path, collecting a monitoring image set and a monitoring audio set, identifying the type of work, and obtaining information distribution of the type of work;

[0008] Perform risk identification based on the distribution of the work type information and obtain risk identification results;

[0009] Integrate the work type information distribution and risk identification results, obtain construction personnel monitoring results, and perform visual display.

[0010] In a second aspect, the present invention provides a construction personnel intelligent management system based on drones, comprising:

[0011] The UAV flight path acquisition module is used to plan the UAV flight path and obtain the flight path within the construction area;

[0012] A work type information distribution module is used to control the drone to fly within the construction area according to the flight path, collect monitoring image sets and monitoring audio sets, identify the work type, and obtain work type information distribution;

[0013] A risk identification module is used to identify risks based on the distribution of the work type information and obtain risk identification results;

[0014] The monitoring visualization display module is used to integrate the work type information distribution and risk identification results, obtain the construction personnel monitoring results, and perform visual display.

[0015] The beneficial effects of the present invention are as follows: the present invention provides an intelligent management method for construction personnel based on drones. Compared with traditional drone applications that mostly remain at the simple image acquisition and video monitoring level, the automated inspection and multimodal recognition of drones realizes intelligent personnel management, and can quickly and accurately identify and locate workers of different types in complex and changeable construction sites, reducing the time and cost of manual inspections and improving management efficiency and accuracy; through the automated drone system, the high manpower requirements of traditional manual inspections are reduced, solving the problems of limited coverage and incomplete information acquisition of traditional methods, significantly reducing human management costs and improving management effects; through real-time group behavior analysis and risk prediction functions, it can dynamically monitor the density and behavior status of personnel, and provide early warning of potential safety risks (such as overcrowding and personnel squeeze). The risk level judgment based on the BIM model enables the system to take timely intervention measures for high-risk areas, significantly improving the safety of the construction site; by generating a three-dimensional visual model of the construction site, it provides managers with an intuitive on-site display, including worker distribution, work status and risk areas, enabling managers to control the overall construction situation in real time, improving decision-making efficiency and the rationality of resource allocation, and effectively solving the current technical problems in drone construction management that are not intelligent, affecting the effectiveness of drone use, and affecting the efficiency and quality of construction safety risk management. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic diagram of a flow chart of a construction personnel intelligent management method based on drones provided by the present invention;

[0017] Figure 2 This is a schematic diagram of the structure of a construction personnel intelligent management system based on drones provided by the present invention;

[0018] Reference numerals: UAV flight path acquisition module 11 , work type information distribution module 12 , risk identification module 13 , monitoring visualization display module 14 . DETAILED DESCRIPTION

[0019] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative efforts shall fall within the scope of protection of the present invention.

[0020] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of the technical features indicated. Therefore, a feature specified as "first" or "second" may explicitly or implicitly include one or more of the specified features. In the description of the present invention, "plurality" means two or more, unless otherwise specifically defined.

[0021] In the description of the present invention, the term "for example" is used to mean "used as an example, illustration or illustration". Any embodiment of the present invention described as "for example" is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any person skilled in the art to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes are not elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed herein.

[0022] Example 1, as Figure 1 As shown, an embodiment of the present invention provides a method for intelligent management of construction personnel based on drones, which specifically includes the following steps:

[0023] S10: Plan the flight path of the UAV in the construction area and obtain the flight path;

[0024] Furthermore, within the construction area, the UAV flight path is planned, and the flight path is obtained by executing step S10, which includes:

[0025] S11: Setting a monitoring start point and a monitoring end point within the construction area;

[0026] S12: According to the monitoring starting point and the monitoring end point, the UAV flight path is planned to obtain the flight path.

[0027] In an embodiment of the present application, the present invention determines the drone detection starting point and end point by pre-setting the drone's monitoring starting point and monitoring end point (used to plan the drone's flight path, such as determined by geographic information system (GIS) data or on-site survey).

[0028] Exemplarily, the method for obtaining the monitoring start point and the monitoring end point is as follows:

[0029] The division of construction areas. Within the construction area, the entire construction site must first be divided to clearly define different construction areas and functional areas. For example, the construction site can be divided into material storage areas, construction operation areas, safety passage areas, etc. The division of these areas can be achieved in the following ways:

[0030] Geographic Information System (GIS) Data: Utilizing the construction site's Geographic Information System (GIS) data, we imported the construction area's boundaries, functional area divisions, and key node locations into the system. GIS data provides precise geographic coordinates, providing the foundation for subsequent route planning.

[0031] Site survey: Through site survey, key locations within the construction area are identified, such as material storage points, construction entrances, safe passages, etc. These locations can be marked using positioning equipment (such as GPS) and their geographic coordinates recorded.

[0032] Determination of the monitoring starting point: The monitoring starting point is the initial position where the UAV begins to perform the monitoring mission.

[0033] In this embodiment, the process of obtaining the monitoring starting point is as follows:

[0034] Based on the construction entrance: Typically, the construction entrance is the primary access point for construction personnel and equipment. Therefore, a safe location near the construction entrance can be used as the monitoring starting point. For example, an open area near the construction entrance can be selected to ensure that the drone can take off safely from this location without interfering with the normal passage of construction personnel and equipment.

[0035] Based on functional area division: If the construction area is large, multiple monitoring starting points can be set according to the functional area division. For example, a starting point can be set in the material storage area, the construction operation area, and the safety passage area respectively, so that the drone can quickly enter each area for monitoring.

[0036] Dynamic Adjustment: In some cases, the layout of the construction area may change (such as the opening of new construction areas or the adjustment of existing areas). In this case, the monitoring starting point can be dynamically adjusted according to the new layout. For example, the construction area division information can be updated in real time through the construction management system, and the monitoring starting point can be re-determined based on this information.

[0037] Determination of the monitoring end point: The monitoring end point is the final location to which the drone returns after completing the monitoring mission. In this embodiment, the process of obtaining the monitoring end point is as follows.

[0038] Same as the starting point: In some embodiments, to simplify the operation process, the monitoring end point can be the same as the monitoring start point. For example, after completing the monitoring mission, the drone returns to the starting point for charging and data transmission.

[0039] Based on the location of the charging station: If a dedicated drone charging station is available within the construction area, the monitoring endpoint can be set at the charging station. The location of the charging station can be determined using GIS data or on-site surveys, ensuring that it maintains a safe distance from other functional areas within the construction area.

[0040] Based on the location of the data processing center: In some complex construction projects, it may be necessary to transmit the data collected by the drone to the data processing center in real time for analysis. In this case, the monitoring end point can be set near the data processing center so that the drone can quickly transmit the data to the processing system.

[0041] Furthermore, according to the monitoring start point and the monitoring end point, the UAV flight path is planned to obtain the flight path, and step S12 includes:

[0042] S121: According to the monitoring start point and the monitoring end point, a rapid expansion random tree algorithm is used to plan the flight path of the UAV to obtain an initial track point;

[0043] S122: performing redundant point trimming on the initial track points to obtain a flight path.

[0044] For example, the rapid expansion random tree algorithm is used to generate initial track points. The specific steps are as follows:

[0045] Algorithm initialization: In the construction area, the RRT algorithm is initialized based on the monitoring starting point (denoted as q_start) and the monitoring ending point (denoted as q_goal). The monitoring starting point is used as the root node of the random tree to construct the initial search tree T.

[0046] Randomly set sampling points and expand the flight path.

[0047] Random sampling: A sampling point q_rand is randomly generated in the free space of the construction area. Based on probability p, the sampling point can be generated randomly or directly point to the target point q_goal with a certain probability to accelerate path convergence.

[0048] Find the nearest node: From the current search tree T, find the node q_near that is closest to the sampling point q_rand.

[0049] Generate a new node: Generate a new node q_new along the direction from q_near to q_rand, and ensure that it meets the step size limit.

[0050] Collision detection: Check whether the path from q_near to q_new collides with obstacles in the construction area (such as buildings, equipment, construction workers, etc.). If the path is obstacle-free, q_new is added to the search tree T.

[0051] Path convergence: Repeat the random sampling and path expansion process until the distance between the newly generated node q_new and the target point q_goal is less than the set threshold. At this point, the target point q_goal is added to the search tree T, and the path planning is completed.

[0052] Get the initial track point: Starting from the target point q_goal, trace back along the parent node until you return to the starting point q_start, and get the initial path from the starting point to the target point. This path consists of a series of track points, which are recorded as the initial track points.

[0053] Pruning redundant points from the initial track: Due to the randomness of the RRT algorithm, the generated initial path may contain a large number of redundant points, which will increase the flight time and energy consumption of the drone. Therefore, it is necessary to prune redundant points from the initial track to optimize the flight path.

[0054] Redundant point judgment: For any two non-adjacent nodes q_i and q_j in the initial track point sequence, if the straight line segment from q_i to q_j does not collide with obstacles in the construction area, all intermediate nodes between these two points are considered redundant points.

[0055] Pruning process: From the initial track point sequence, check each pair of non-adjacent nodes q_i and q_j in turn.

[0056] For node pairs that meet the redundancy condition, delete all intermediate nodes between q_i and q_j.

[0057] Repeat the above process until no more redundant points can be found.

[0058] Final flight path: After trimming redundant points, the optimized flight path is obtained.

[0059] The flight path obtained through the above embodiment solves the problems of high search randomness, redundant paths and poor path smoothness. Under the premise of ensuring the safe flight of the UAV, it achieves the technical effect of reducing the number of track points, reducing flight time and energy consumption.

[0060] S20: Controlling the UAV to fly within the construction area according to the flight path, collecting a monitoring image set and a monitoring audio set, performing work type identification, and obtaining work type information distribution;

[0061] Furthermore, the UAV is controlled to fly within the construction area according to the flight path, a monitoring image set and a monitoring audio set are collected, the type of work is identified, and the type of work information distribution is obtained. Step S20 includes:

[0062] S21: Controlling the UAV to fly within the construction area according to the flight path, collecting monitoring images and monitoring audio at multiple locations, and obtaining a monitoring image set and a monitoring audio set;

[0063] Furthermore, flight path control: The drone flies within the construction area according to a pre-planned flight path. This flight path is generated using a rapidly expanding random tree (RRT) algorithm and optimized by redundant point pruning to ensure the drone can efficiently cover all key locations in the construction area.

[0064] First, the monitoring data is collected, which specifically includes the following contents.

[0065] During the flight, the drone collected monitoring images and audio data at multiple locations.

[0066] Monitoring image acquisition: Drones equipped with high-definition cameras capture images of the construction area at fixed intervals or by triggering at specific locations. This image data forms a monitoring image collection.

[0067] Monitoring Audio Collection: The drone is equipped with an audio sensor to collect sound signals from the construction area. These sound signals are processed for noise reduction and then converted into monitoring audio data.

[0068] S22: training the job recognition model;

[0069] S23: Inputting the monitoring image set and the monitoring audio set into the work type recognition model, and outputting a work type information set and a location information set;

[0070] S24: Generate a job type information distribution based on the job type information set and the location information set.

[0071] Furthermore, the job type recognition model is trained, and step S22 is performed including:

[0072] S221: Collecting a sample monitoring image set and a sample monitoring audio set, and collecting work type information and location information in different sample monitoring images and sample monitoring audios to obtain a sample work type information set and a sample location information set;

[0073] S222: Build a job recognition model based on YOLOv5s and MFCC;

[0074] S223: Preprocess the sample monitoring image set and the sample monitoring audio set, and train the job type recognition model in combination with the sample job type information set and the sample location information set until the requirements are met to obtain a job type recognition model.

[0075] For example, assume the construction site is a small building site, and the construction area is a rectangular area with a length of 100 meters and a width of 50 meters. The following types of work are performed within the construction area:

[0076] Electricians: 20 people, concentrated near the distribution box (coordinate range: [20,30] to [30,40]);

[0077] Welders: 15 people, concentrated in the steel structure welding area (coordinate range: [60,10] to [70,20]);

[0078] Scaffolders: 10 people, distributed in the scaffolding construction area (coordinate range: [80,30] to [90,40]).

[0079] Data collection and preprocessing are performed as follows.

[0080] Data collection sample monitoring image collection: Use high-definition cameras to collect image data from different locations within the construction area.

[0081] Each image sample contains the appearance features of the construction workers (such as safety helmets, work clothes, tools, etc.).

[0082] Sample monitoring audio collection: Audio sensors are used to collect audio data from different locations within the construction area.

[0083] Audio samples include construction workers' voice instructions, tool operation sounds, etc.

[0084] Work type information and location information: Obtain the work type information and location coordinates corresponding to each sample through manual labeling or on-site recording.

[0085] For example: the electrician is at [22,32], the welder is at [65,15], and the scaffolder is at [85,35].

[0086] Furthermore, data preprocessing is performed.

[0087] Image preprocessing: Perform noise reduction and normalization on the image data. Use the OpenCV library to crop the image to a fixed size (e.g., 224×224 pixels).

[0088] Audio preprocessing: Perform noise reduction on audio data.

[0089] Extract the Mel-Frequency Cepstral Coefficient (MFCC) features of the audio for subsequent audio analysis.

[0090] Furthermore, the job identification model is constructed and trained.

[0091] Model construction, build a job identification model based on YOLOv5s and MFCC.

[0092] Image recognition module: Uses the YOLOv5s network structure to detect and classify targets in image data and identify the appearance characteristics of construction workers.

[0093] Audio recognition module: Extracts MFCC features of audio and uses recurrent neural networks (RNN) or long short-term memory networks (LSTM) for audio classification.

[0094] Multimodal fusion module: fuses image and audio features and outputs the final job recognition result through a fully connected layer.

[0095] Model training and dataset preparation: The collected image and audio samples are divided into training sets and validation sets. The training set is used for model training, and the validation set is used to evaluate model performance.

[0096] Training process: Use labeled sample data to train the model.

[0097] Data enhancement techniques (such as image rotation, scaling, audio noise injection, etc.) are used to improve the generalization ability of the model.

[0098] Through cross-validation and hyperparameter optimization, the high accuracy of the model is ensured in different construction scenarios.

[0099] The job type information distribution is generated as follows:

[0100]

[0101] Model inference: The collected monitoring image and audio collections are input into the trained work type recognition model, and the model outputs the work type information and location coordinates of each position.

[0102] The type of work information distribution map is generated based on the type of work information and location information output by the model, and the type of work information set and location information set are output.

[0103] S30: Perform risk identification based on the job type information distribution to obtain a risk identification result.

[0104] Furthermore, based on the distribution of the job information, risk identification is performed to obtain a risk identification result. Step S30 includes:

[0105] S31: extracting and obtaining worker location distribution based on the job type information distribution;

[0106] S32: Calculate the worker density distribution based on the worker position distribution;

[0107] S33: Predicting the construction area risk based on the worker density distribution as a risk identification result.

[0108] For example, assume that the construction site is a medium-sized construction site, and the construction area is a rectangular area with a length of 200 meters and a width of 100 meters. The following types of work are performed within the construction area:

[0109] Electricians: 30 people, concentrated near the distribution box (coordinate range: [30,40] to [50,60]);

[0110] Welders: 20 people, concentrated in the steel structure welding area (coordinate range: [120,20] to [140,40]);

[0111] Scaffolders: 15 people, distributed in the scaffolding construction area (coordinate range: [160,60] to [180,80]);

[0112] Safety supervisors: 5 persons, distributed at key locations in the construction area.

[0113] The risk identification acquisition process is as follows.

[0114] Extracting Worker Location Distribution: Based on the job type distribution, extract each worker's location information. This location information is output by the job type recognition model and contains the specific coordinates of each worker.

[0115] The extracted worker location information is shown in the following table.

[0116] Type of work X coordinate Y coordinate electrician 32 45 electrician 40 55 welder 125 25 welder 130 35 Scaffolder 165 65 Scaffolder 170 75 Safety Supervisor 80 50 Safety Supervisor 100 30 ... ... ...

[0117] To calculate the worker density distribution, the construction area is divided into multiple small grids (for example, each grid is 10 meters × 10 meters in size), the number of workers in each grid is counted, and the worker density distribution is calculated.

[0118] Grid division: The construction area is 200 meters × 100 meters, divided into 20 × 10 grids, each grid is 10 meters × 10 meters; the grids are numbered from (1,1) to (20,10) starting from the lower left corner.

[0119] Density calculation: Count the number of workers in each grid; calculate the worker density of each grid (number of workers / grid area).

[0120] The following is example data for worker density distribution.

[0121] Grid Number Number of workers Worker density (people / 100 square meters) (3,4) 5 0.5 (4,5) 6 0.6 (12,2) 4 0.4 (13,3) 5 0.5 (16,6) 3 0.3 (17,7) 4 0.4 ... ... ...

[0122] Predict construction area risks. Based on the density of workers, predict the risks in the construction area. Risk prediction can be based on the following rules.

[0123] Low-risk areas: low worker density (<0.3 people / 100 square meters).

[0124] Medium-risk area: medium worker density (0.3-0.6 people / 100 square meters).

[0125] High-risk areas: high worker density (>0.6 people / 100 square meters).

[0126] S40: Integrate the work type information distribution and risk identification results, obtain construction personnel monitoring results, and perform visual display.

[0127] Furthermore, the work type information distribution and risk identification results are integrated to obtain construction personnel monitoring results for visualization. Step S40 includes:

[0128] S41: Integrate the work type information distribution and risk identification results, combine with the BIM model of the construction area, and generate construction personnel monitoring results;

[0129] S42: Transmitting the construction worker monitoring results to a data center for storage, and sending them to a terminal for visual display.

[0130] For example, a BIM model (Building Information Modeling) is introduced. A BIM model is a three-dimensional digital model of the construction area, containing information such as the building structure, equipment layout, and construction progress. By combining the distribution of work type information and risk identification results with the BIM model, a more intuitive visualization can be achieved.

[0131] Data integration: Integrate the work type information distribution and risk identification results into the BIM model to generate construction worker monitoring results. The specific steps are as follows.

[0132] Data mapping: Map the type of work information and risk level to the corresponding location in the BIM model.

[0133] 3D visualization: Use the 3D view of the BIM model to display the location of construction workers, job type information, and risk level.

[0134] Data transmission and storage: The integrated monitoring results of construction workers are transmitted to the data center for storage and visualized through terminal devices. The terminal device can be a computer, tablet or large on-site monitoring screen.

[0135] For visual interface design, BIM software (such as Revit, Navisworks) or WebGL technology can be used to achieve three-dimensional visualization.

[0136] The embodiment of the present invention provides a method for intelligent management of construction personnel based on drones, which has at least the following technical effects:

[0137] Compared to traditional drone applications, which mostly remain at the level of simple image acquisition and video monitoring, the embodiments of the present invention achieve intelligent personnel management through the automated inspection and multimodal recognition of drones. They can quickly and accurately identify and locate workers of different trades in complex and changing construction sites, reducing the time and cost of manual inspections and improving management efficiency and accuracy. The automated drone system reduces the high manpower requirements of traditional manual inspections, solves the problems of limited coverage and incomplete information acquisition of traditional methods, significantly reduces labor management costs and improves management effectiveness. Through real-time group behavior analysis and risk prediction functions, it can dynamically monitor the density and behavior of personnel, and provide early warning of potential safety risks (such as overcrowding and personnel squeeze). Based on the risk level judgment of the BIM model, the system enables timely intervention measures for high-risk areas, significantly improving the safety of the construction site. By generating a three-dimensional visualization model of the construction site, it provides managers with an intuitive on-site display, including worker distribution, work status and risk areas, enabling managers to have real-time control over the overall construction situation, improving decision-making efficiency and the rationality of resource allocation, and effectively solving the technical problems of the current lack of intelligence in drone construction management, which affects the effectiveness of drone use.

[0138] Example 2, as Figure 2 As shown, based on the same inventive concept as the intelligent management method for construction personnel based on drones provided in Example 1, an embodiment of the present invention further provides an intelligent management system for construction personnel based on drones. The explanation of the intelligent management method for construction personnel based on drones in Example 1 is also applicable to an intelligent management system for construction personnel based on drones, which includes:

[0139] The UAV flight path acquisition module 11 is used to plan the UAV flight path in the construction area and obtain the flight path;

[0140] A work type information distribution module 12 is configured to control the UAV to fly within the construction area according to the flight path, collect monitoring image sets and monitoring audio sets, identify work types, and obtain work type information distribution;

[0141] The risk identification module 13 is used to identify risks based on the distribution of the job information and obtain risk identification results;

[0142] The monitoring visualization display module 14 is used to integrate the work type information distribution and risk identification results, obtain the construction personnel monitoring results, and perform visualization display.

[0143] Furthermore, the UAV flight path acquisition module 11 is also used to:

[0144] In the construction area, set the monitoring starting point and monitoring ending point;

[0145] According to the monitoring starting point and the monitoring end point, the UAV flight path planning is performed to obtain the flight path.

[0146] Furthermore, the UAV flight path acquisition module 11 is also used to:

[0147] According to the monitoring starting point and monitoring end point, the UAV flight path is planned using the rapid expansion random tree algorithm to obtain the initial track point;

[0148] Redundant points are trimmed from the initial track points to obtain a flight path.

[0149] Furthermore, the job type information distribution module 12 is also used to:

[0150] Controlling the drone to fly within the construction area according to the flight path, collecting monitoring images and monitoring audio at multiple locations, and obtaining a monitoring image set and a monitoring audio set;

[0151] Training job recognition models;

[0152] Input the monitoring image set and the monitoring audio set into the work type recognition model, and output a work type information set and a location information set;

[0153] Based on the job type information set and the location information set, a job type information distribution is generated.

[0154] Furthermore, the job type information distribution module 12 is also used to:

[0155] Collecting a sample monitoring image set and a sample monitoring audio set, and collecting work type information and location information in different sample monitoring images and sample monitoring audios to obtain a sample work type information set and a sample location information set;

[0156] Build a job recognition model based on YOLOv5s and MFCC;

[0157] The sample monitoring image set and the sample monitoring audio set are preprocessed, and the job type recognition model is trained in combination with the sample job type information set and the sample location information set until the requirements are met to obtain a job type recognition model.

[0158] Furthermore, the risk identification module 13 is further configured to:

[0159] Extracting and obtaining worker location distribution based on the job type information distribution;

[0160] Calculating worker density distribution based on the worker position distribution;

[0161] Based on the worker density distribution, the construction area risk is predicted and obtained as a risk identification result.

[0162] Furthermore, the monitoring visualization display module 14 is also used to:

[0163] Integrate the work type information distribution and risk identification results, combine with the BIM model of the construction area, and generate construction personnel monitoring results;

[0164] The construction worker monitoring results are transmitted to the data center for storage and sent to the terminal for visual display.

[0165] It should be noted that, in the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0166] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0167] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts 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, a special-purpose computer, an embedded computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0168] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0169] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0170] Although preferred embodiments of the present invention have been described, additional changes and modifications to these embodiments may occur to those skilled in the art once the basic inventive concepts become known.

[0171] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the present invention and its equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A construction personnel intelligent management method based on drones, characterized in that: The method comprises: In the construction area, the UAV flight path is planned to obtain the flight path; Controlling the drone to fly within the construction area according to the flight path, collecting a monitoring image set and a monitoring audio set, identifying the type of work, and obtaining information distribution of the type of work; Perform risk identification based on the distribution of the work type information and obtain risk identification results; Integrate the work type information distribution and risk identification results, obtain construction personnel monitoring results, and perform visual display.

2. The intelligent management method for construction personnel based on drones according to claim 1 is characterized in that: In the construction area, the UAV flight path is planned to obtain the flight path, including: In the construction area, set the monitoring starting point and monitoring ending point; According to the monitoring starting point and the monitoring end point, the UAV flight path planning is performed to obtain the flight path.

3. The intelligent management method for construction personnel based on drones according to claim 2 is characterized in that: According to the monitoring starting point and monitoring end point, the UAV flight path is planned to obtain the flight path, including: According to the monitoring starting point and monitoring end point, the UAV flight path is planned using the rapid expansion random tree algorithm to obtain the initial track point; Redundant points are trimmed from the initial track points to obtain a flight path.

4. The intelligent management method for construction personnel based on drones according to claim 1 is characterized in that: Controlling the drone to fly within the construction area according to the flight path, collecting a monitoring image set and a monitoring audio set, performing work type identification, and obtaining work type information distribution, including: Controlling the drone to fly within the construction area according to the flight path, collecting monitoring images and monitoring audio at multiple locations, and obtaining a monitoring image set and a monitoring audio set; Training job recognition models; Input the monitoring image set and the monitoring audio set into the work type recognition model, and output a work type information set and a location information set; Based on the job type information set and the location information set, a job type information distribution is generated.

5. The intelligent management method for construction personnel based on drones according to claim 4 is characterized in that: Training the job recognition model, including: Collecting a sample monitoring image set and a sample monitoring audio set, and collecting work type information and location information in different sample monitoring images and sample monitoring audios to obtain a sample work type information set and a sample location information set; Build a job recognition model based on YOLOv5s and MFCC; The sample monitoring image set and the sample monitoring audio set are preprocessed, and the job type recognition model is trained in combination with the sample job type information set and the sample location information set until the requirements are met to obtain a job type recognition model.

6. The intelligent management method for construction personnel based on drones according to claim 1 is characterized in that: Based on the distribution of the job information, risk identification is performed to obtain risk identification results, including: Extracting and obtaining worker location distribution based on the job type information distribution; Calculating worker density distribution based on the worker position distribution; Based on the worker density distribution, the construction area risk is predicted and obtained as a risk identification result.

7. The intelligent management method for construction personnel based on drones according to claim 1 is characterized in that: Integrate the work type information distribution and risk identification results to obtain construction worker monitoring results and perform visual display, including: Integrate the work type information distribution and risk identification results, combine with the BIM model of the construction area, and generate construction personnel monitoring results; The construction worker monitoring results are transmitted to the data center for storage and sent to the terminal for visual display.

8. An intelligent management system for construction personnel using drones, characterized in that: The system is used to perform the method according to any one of claims 1 to 7, and the system includes: The UAV flight path acquisition module is used to plan the UAV flight path and obtain the flight path within the construction area; A work type information distribution module is used to control the drone to fly within the construction area according to the flight path, collect monitoring image sets and monitoring audio sets, identify the work type, and obtain work type information distribution; A risk identification module is used to identify risks based on the distribution of the work type information and obtain risk identification results; The monitoring visualization display module is used to integrate the work type information distribution and risk identification results, obtain the construction personnel monitoring results, and perform visual display.

Images