Construction site safety management method and system integrated with head-mounted equipment

Abstract

The invention provides a construction site safety management method and system integrated with a head-mounted device, and the method comprises the steps: collecting a construction site image through the head-mounted device, transmitting the construction site image to a server, carrying out the target scene recognition based on the construction site image, carrying out the operation recognition of the construction site image based on a preset recognition strategy for the target scene, and obtaining a target scene recognition result. Whether the construction site meets the construction safety specification is determined, automatic construction site safety management is achieved, and the construction site safety management efficiency is improved. And meanwhile, the safety detection result of the construction site is sent to the preset client, so that related personnel can conveniently know the safety condition of the construction site. Besides, the head-mounted equipment is used in the construction site to collect the construction site image for identification, so that the construction site image can comprise a part which is difficult to arrive manually, the safety inspection range of the construction site is expanded, and the comprehensiveness of safety management of the construction site and the construction safety are further improved.

Description

Technical Field

[0001] The present invention relates to the technical field of safety management, and particularly to a construction site safety management method and system integrating a head-mounted device. Background Art

[0002] Safety is of utmost importance in construction projects. Each construction enterprise usually formulates a large number of safety management regulations to ensure the safety of the construction site. However, there is a phenomenon that there are few personnel in the project safety department in the traditional construction project safety management chain. It is difficult to comprehensively cover the production work through the safety department personnel to supervise the construction site. At the same time, the safety management capabilities of different personnel vary, the supervision effectiveness is insufficient, and there is a lack of necessary knowledge reserves for discovering potential hazards and avoiding risks; there are blind spots in responsibilities in the actual management process. In addition, a certain proportion of the project safety department not only needs to supervise the on-site work, but also daily check for potential hazards and implement the rectification of corresponding potential hazards. The safety awareness of on-site technicians is more degraded, resulting in management blind spots in on-site production safety, and accident risks are exposed.

[0003] All of the above factors lead to low management efficiency and poor effects in the current construction site management, which greatly threatens the construction safety of construction projects. Summary of the Invention

[0004] In view of this, embodiments of the present invention provide a construction site safety management method and system integrating a head-mounted device to improve the management efficiency and safety of the construction site.

[0005] According to one aspect of the present invention, there is provided a construction site safety management method integrating a head-mounted device, which is applied to a server, and the method includes:

[0006] Obtain an image of the construction site collected by the head-mounted device as a target image to be recognized;

[0007] Input the target image into a preset scene recognition model, and obtain a scene recognition result output by the preset scene recognition model based on the target image as a target scene, where the target scene includes at least fixed-point inspection, high-altitude operation, hot work operation, or confined space operation;

[0008] Based on a safety recognition strategy preset for the target scene, recognize the target image to obtain a safety detection result of the target image, where the safety recognition strategy includes one or more of safety helmet recognition, electrical box recognition, hot work operation recognition, reflective vest recognition, high-altitude operation safety belt recognition, and edge protection detection;

[0009] Send the safety detection result to a preset client so that the preset client displays the safety detection result.

[0010] In a possible embodiment, the method further includes: when the security detection result does not meet the security specifications, determining the risk level corresponding to the security detection result according to a preset risk assessment standard; wherein, the preset risk assessment standard includes risk severity, occurrence probability, and scope of influence;

[0011] Sending the risk level of the security detection result to the preset client, so that the preset client displays the risk level.

[0012] Before obtaining the construction site image collected by the head-mounted device as the target image to be recognized, the method further includes:

[0013] Receiving a task initiation request, wherein the task initiation request includes a target task type and initiator information, and the target task type includes safety inspection, daily inspection, dangerous operation application, and safety acceptance;

[0014] Sending the initiator information to a preset approval client, so that the preset approval client approves the task initiation request based on the user information corresponding to the target task type; wherein, there are multiple preset approval clients;

[0015] When all the preset approval clients approve the task initiation request, allowing the initiator to execute the target task; and executing the step of obtaining the construction site image collected by the head-mounted device as the target image to be recognized during the execution of the target task;

[0016] If there is a preset approval client that does not approve the task initiation request, rejecting the initiator from executing the target task.

[0017] In a possible embodiment, the server includes a deduction large model, and the method further includes: obtaining the audio collected by the head-mounted device;

[0018] Extracting the problem information contained in the audio and outputting the problem information to the deduction large model;

[0019] Obtaining the Q&A result output by the deduction large model and returning it to the head-mounted device.

[0020] In a possible embodiment, the method further includes: regularly sending inspection tasks to the clients with the inspection task permissions according to a preset inspection cycle, and simultaneously sending the inspection list included in the inspection tasks.

[0021] According to another aspect of the present invention, a construction site safety management system integrating a head-mounted device is provided. The system includes a head-mounted device and a server. Among them, the head-mounted device is used to collect construction site images and send the construction site images to the server;

[0022] The server is used to input the construction site image as a target image into a preset scene recognition model, and obtain the scene recognition result output by the preset scene recognition model based on the target image as the target scene. Among them, the target scene at least includes fixed-point inspection, high-altitude operation, hot work operation or confined space operation;

[0023] Based on the safety recognition strategy preset for the target scene, the target image is recognized to obtain the safety detection result of the target image. Among them, the safety recognition strategy includes one or more of safety helmet recognition, electrical box recognition, hot work operation recognition, reflective vest recognition, high-altitude operation safety belt recognition and edge protection detection;

[0024] Send the safety detection result to a preset client so that the preset client displays the safety detection result.

[0025] In a possible embodiment, a safety helmet recognition model and a reflective vest recognition model are integrated in the head-mounted device; the head-mounted device is used to input the construction site image into the safety helmet recognition model to identify whether the safety helmet is worn in compliance in the construction site image, and / or input the construction site image into the reflective vest recognition model to identify whether the reflective vest is worn in compliance in the construction site image.

[0026] In a possible embodiment, the server is used to determine the risk level corresponding to the safety detection result according to a preset risk assessment standard when the safety detection result does not meet the safety specifications; among them, the preset risk assessment standard includes risk severity, occurrence probability and influence range;

[0027] Send the risk level of the safety detection result to the preset client so that the preset client displays the risk level;

[0028] Before obtaining the construction site image collected by the head-mounted device as the target image to be recognized, the method further includes:

[0029] Receive a task initiation request, where the task initiation request includes a target task type and initiator information, and the target task type includes safety inspection, daily inspection, dangerous operation application and safety acceptance;

[0030] Send the initiator information to a preset approval client, so that the preset approval client approves the task initiation request based on the user information corresponding to the target task type; wherein, there are multiple preset approval clients;

[0031] When each of the preset approval clients approves the task initiation request, allow the initiator to execute the target task; and execute the step of acquiring the construction site image collected by the head-mounted device during the execution of the target task as the target image to be recognized;

[0032] If there is a preset approval client that does not approve the task initiation request, reject the initiator from executing the target task;

[0033] The server includes a deduction large model, and the head-mounted device is further used to collect construction site audio and send the audio to the server;

[0034] The server is used to extract the problem information included in the audio and output the problem information to the deduction large model;

[0035] Obtain the Q&A result output by the deduction large model and return it to the head-mounted device;

[0036] The server is also used to regularly send inspection tasks to the client with the inspection task permission according to a preset inspection period, and simultaneously send the inspection list included in the inspection task.

[0037] According to another aspect of the present invention, there is provided an electronic device, including:

[0038] A processor; and

[0039] A memory storing a program,

[0040] wherein, the program includes instructions, and when the instructions are executed by the processor, the processor executes any one of the above-mentioned construction site safety management methods of the integrated head-mounted device.

[0041] According to another aspect of the present invention, there is provided a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause a computer to execute any one of the above-mentioned construction site safety management methods of the integrated head-mounted device.

[0042] One or more technical solutions provided in the embodiments of the present invention collect construction site images through a head-mounted device and send them to a server. The server performs construction scene recognition based on the construction site images. After determining the target scene corresponding to the construction site image, based on the preset recognition strategy for the target scene, the construction site image is subjected to operation recognition to determine whether the construction site complies with construction safety specifications, realizing automated construction site safety management and improving the efficiency of construction site safety management. At the same time, by sending the target recognition result of the construction site to a preset client, the construction site safety inspection result can reach more people, facilitating the relevant personnel to understand the safety situation of the construction site. In addition, by using a head-mounted device at the construction site to collect and recognize construction site images, the construction site images can include parts that are difficult for humans to reach, expanding the scope of construction site safety inspection, and thus providing the comprehensiveness of construction site safety management and construction safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In the following description of the exemplary embodiments in conjunction with the drawings, more details, features, and advantages of the present invention are disclosed. In the drawings:

[0044] Figure 1 FIG. is a flowchart of a method for construction site safety management integrating a head-mounted device provided by an embodiment of the present invention;

[0045] Figure 2 FIG. is a flowchart of the use of a construction site safety management software integrating a head-mounted device provided by an embodiment of the present invention;

[0046] Figure 3 FIG. is a schematic architecture diagram of a construction site safety management system integrating a head-mounted device provided by an embodiment of the present invention;

[0047] Figure 4 FIG. shows a structural block diagram of an exemplary electronic device that can be used to implement the embodiments of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0048] The embodiments of the present invention will be described in more detail below with reference to the drawings. Although some embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. On the contrary, these embodiments are provided to more thoroughly and completely understand the present invention. It should be understood that the drawings and embodiments of the present invention are only for exemplary purposes and are not used to limit the protection scope of the present invention.

[0049] It should be understood that the various steps described in the method embodiments of the present invention can be executed in different orders and / or executed in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this regard.

[0050] As used herein, the term "including" and its variants are open-ended, that is, "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description. It should be noted that the concepts such as "first" and "second" mentioned in the present invention are only used to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependent relationships.

[0051] It should be noted that the modifications of "one" and "plural" mentioned in the present invention are illustrative rather than restrictive. Those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0052] The names of the messages or information exchanged between multiple devices in the embodiments of the present invention are only for illustrative purposes and are not used to limit the scope of these messages or information.

[0053] In order to improve the efficiency and safety of construction site safety management, the embodiments of the present invention provide a construction site safety management method and system integrating a head-mounted device. The construction site safety management method provided by the embodiments of the present invention can be applied to any electronic device with construction site safety management functions, and the electronic device can be a server, a computer or a mobile terminal, etc. The solution of the present invention will be described below with reference to the drawings:

[0054] In a possible embodiment, the construction site safety management method integrating a head-mounted device provided by the embodiments of the present invention can be applied to the server side. Among them, the server side is a software program running on the server, which is used to process requests from the client and provide corresponding services, such as Figure 1 as shown, Figure 1 is a schematic flowchart of a construction site safety management method integrating a head-mounted device provided by the embodiments of the present invention, and may include the following steps:

[0055] S101. Obtain the construction site image collected by the head-mounted device as the target image to be recognized;

[0056] S102. Input the target image into a preset scene recognition model, and obtain the scene recognition result output by the preset scene recognition model based on the target image as the target scene, where the target scene includes at least fixed-point inspection, high-altitude operation, hot work, or confined space operation;

[0057] S103. Based on the safety recognition strategy preset for the target scene, recognize the target image to obtain the safety detection result of the target image, where the safety recognition strategy includes one or more of safety helmet recognition, electrical box recognition, hot work recognition, reflective vest recognition, high-altitude operation safety belt recognition, and edge protection detection;

[0058] S104. Send the safety detection result to a preset client so that the preset client can display the safety detection result.

[0059] Applying the embodiments of the present invention, the construction site image is collected by a head-mounted device and sent to the server. The server performs construction scene recognition based on the construction site image. After determining the target scene corresponding to the construction site image, based on the preset recognition strategy for the target scene, the construction site image is recognized for operations to determine whether the construction site complies with construction safety specifications, realizing automated construction site safety management and improving the efficiency of construction site safety management. At the same time, by sending the target recognition result of the construction site to a preset client, the safety inspection result of the construction site can reach more people, facilitating the relevant personnel to understand the safety situation of the construction site. In addition, by using a head-mounted device at the construction site to collect and recognize the construction site image, the construction site image can include parts that are difficult for humans to reach, expanding the scope of construction site safety inspection, and thus providing the comprehensiveness of construction site safety management and construction safety.

[0060] The following is an exemplary description of the above S101 - S104:

[0061] In S101, the construction site images can be collected through the head-mounted devices worn by the personnel at the construction site. As an advanced wearable technology, the head-mounted device can provide real-time information feedback and interaction capabilities for the workers and managers at the construction site. Exemplarily, the head-mounted device may include components such as a camera, a microphone, a display screen, and sensors. Among them, the camera is used to collect construction site images. Exemplarily, relevant personnel can collect construction site images by touching the screen or pressing mechanical buttons, etc. The content included in the construction site images can be set according to the actual application scenarios, such as it can include the building under construction, construction personnel, etc. The microphone is used to collect the audio of the construction site personnel and can communicate with remote personnel through a wireless network; the display screen is used to display the collected images or the information received from the server. The above sensors may include a temperature sensor and a humidity sensor, which are used to collect the environmental information of the construction site. Through the head-mounted device, when on-site workers encounter problems, they can have a video call with remote experts in real time, obtain immediate guidance and support, improve the efficiency of problem-solving, and also greatly reduce the safety hazards caused by decision-making delays.

[0062] In a possible embodiment, the construction site images collected by the head-mounted device can be uploaded to the server through a wireless network, and the server further identifies the construction site images. For the convenience of narration, the construction site images collected by the head-mounted device will be referred to as target images in the following text.

[0063] After receiving the target image, the server can output the target image to a preset scene recognition model to obtain the scene recognition result output by the preset scene recognition model for the target image. The above-mentioned preset scene recognition model can be trained from models such as CNN (Convolutional Neural Network) and RNN (Recurrent Neural Network). As a possible implementation, a preset scene training dataset can be set. The scene training dataset can include multiple scene training images, and each scene training image has been pre-annotated with a scene, that is, each scene training image corresponds to a specific scene label, and the scene label can include fixed-point inspection, hot work, high-altitude work, and confined space work, etc. Among them, fixed-point inspection refers to conducting safety inspections at fixed locations. For example, at a construction site, regular safety inspections may be carried out at fixed locations such as the entrance of the construction site and the construction site office. Hot work refers to non-conventional operations that use open flames or may generate sparks or hot surfaces at the construction site, such as operations carried out with tools such as welding, cutting, heat treatment, blowtorches, electric drills, and grinders. Confined space work refers to work carried out in enclosed or partially enclosed spaces, which usually have small entrances and exits, poor ventilation, and may contain hazardous substances such as toxic and harmful gases and flammable and explosive substances. For example, it can include work in places such as basements, tunnels, pipelines, ship holds, storage tanks, and containers.

[0064] Input each scene training image into the initial scene recognition model. The scene recognition model can perform image segmentation, feature extraction, etc. on the scene training image to obtain the scene prediction result corresponding to the scene training image, calculate the difference between the scene prediction result of each scene training image and the scene label based on a preset loss function, and adjust the parameters of the scene recognition model based on this difference until the difference converges to obtain a trained scene recognition model. The above loss function can be a mean square error loss function, a cross-entropy loss function, etc.

[0065] After obtaining the trained scene recognition model, the scene recognition model can be stored in the server. Exemplarily, each network layer included in the scene recognition model and the parameters of the corresponding network layer can be stored. In this way, after the server receives the target image, the target image can be input into the scene recognition model to obtain the scene recognition result of the target image.

[0066] In a possible embodiment, security recognition policies are preset for different scenarios. The security recognition policies can be set according to a preset security manual within an enterprise. Exemplarily, in spot checks, it is necessary to check whether construction site personnel are wearing safety helmets, reflective vests, and whether protective nets are installed on the on-site buildings. Therefore, the security recognition policy corresponding to the spot check scenario can include safety helmet recognition, reflective vest recognition, and edge protection detection. For another example, in high-altitude operations, it is necessary to check whether construction site personnel are wearing safety helmets, safety belts, reflective vests, etc. Therefore, the security recognition policy corresponding to the high-altitude operation scenario can include safety helmet recognition, high-altitude operation safety belt recognition, reflective vest recognition, and edge protection detection, etc. The above security recognition policies can also include electrical box recognition and hot work operation recognition. Among them, electrical box recognition is used to identify whether the electrical boxes at the construction site are connected properly, such as whether the electrical box doors are closed and whether the electrical box wiring is normal. Hot work operation recognition can include identifying whether relevant personnel are wearing protective gloves, protective helmets, etc.

[0067] The above security recognition policies can be implemented through a pre-trained recognition model. Exemplarily, the recognition model can include a safety helmet recognition model, an electrical box recognition model, a hot work operation recognition model, a reflective vest recognition model, a high-altitude operation safety belt recognition model, and an edge protection detection model, etc. Taking the training method of the safety helmet recognition model as an example, the training process of each recognition model will be exemplarily described below:

[0068] In a possible embodiment, safety helmet training images can be obtained. Each safety helmet training image includes images of personnel wearing safety helmets and images of personnel not wearing safety helmets. Each safety helmet training image has been pre-annotated. Specifically, images containing personnel wearing safety helmets are annotated as "yes", and images containing personnel not wearing safety helmets are annotated as "no". Input each safety helmet training image into an initial safety helmet training model, so that the initial safety helmet recognition model performs image segmentation and feature extraction on each safety helmet training image, and outputs a safety helmet wearing result based on the extracted features. The above initial safety helmet recognition model can be a CNN model, an RNN model, etc. Calculate the difference between the safety helmet wearing result and the safety helmet wearing label of the image through a preset loss function, and adjust the parameters of the initial safety helmet recognition model based on this difference through methods such as the gradient descent method until the difference converges to obtain a trained safety helmet recognition model. The convergence of this difference means that the difference is less than a preset difference threshold or the difference between the differences obtained from two trainings is less than a preset difference threshold.

[0069] In a possible embodiment, after training the above-mentioned safety helmet recognition model, electrical box recognition model, hot work operation recognition model, reflective vest recognition model, high-altitude work safety belt recognition model, and edge protection detection model, the model parameters can be stored in the server corresponding to the recognition model identifiers. The model parameters can include each network layer included in the model and the parameters of the network layer.

[0070] In a possible embodiment, the server may further include the correspondence between the scene identifier and the safety policy identifier. One safety policy identifier can correspond to one or more safety recognition models. After identifying the target scene corresponding to the target image through the scene recognition model, the corresponding safety policy can be obtained based on the target scene identifier, and the target image can be recognized through the safety recognition models included in the safety policy to obtain the safety detection result of the target image.

[0071] Since the head-mounted device is actually a microcomputer, the safety recognition models with less computing power consumption can be stored in the head-mounted device, and the construction site images can be recognized through the head-mounted device. Exemplarily, the safety helmet and safety belt recognition models can be deployed in the head-mounted device for on-site recognition.

[0072] In a possible embodiment, it is also possible to determine whether the safety detection result is compliant based on the safety rules corresponding to each scene, and add the result of compliance or non-compliance to the safety detection result, so that relevant personnel can intuitively understand whether the construction site meets the safety specifications.

[0073] In a possible embodiment, the safety assessment of the detection result can also be carried out according to the preset risk assessment criteria. Exemplarily, the above-mentioned risk assessment criteria can include the risk severity, occurrence probability, and impact range. Exemplarily, in the case where the detection result does not conform to the safety rules corresponding to the scene, the risk severity can be determined based on the economic loss caused by the detection result, the risk occurrence probability corresponding to the detection result can be determined based on the historical detection result records, the impact range of the detection result can be determined based on the number of links affected by the detection result, and the risk level corresponding to the detection result can be determined based on the sum of each index, and the risk level can be added to the safety detection result.

[0074] After obtaining the above safety detection result, the above safety detection result can be sent to a preset client, which corresponds to the user. Exemplarily, the preset client can be set according to the client identifier of the person in charge of the construction site as preset, so that the person in charge can obtain the safety detection result of the construction site in a timely manner.

[0075] In a possible embodiment, the preset client can take corresponding measures according to whether the detection result is compliant and the risk level. For example, it can stop the construction site operation or notify the corresponding personnel to wear safety helmets, safety belts, etc. As a possible implementation, improvement suggestions can be preset according to the risk type, and the safety detection results and the corresponding improvement suggestions can be sent to the preset client to improve the risk management efficiency.

[0076] The above-mentioned server can also monitor the improvement situation of the risk. Exemplarily, the corresponding relationship between the risk improvement measures and the users can be preset. After the user executes the risk improvement measures, the improvement result can be uploaded to the server. If the risk improvement result is not received for a long time, the server can push the risk improvement task to the corresponding client of the user to improve the risk improvement efficiency.

[0077] In a possible embodiment, the above method may further include:

[0078] S111. Receive a task initiation request, where the task initiation request includes a target task type and initiator information, and the target task type includes safety inspection, daily inspection, dangerous operation application, and safety acceptance;

[0079] S112. Send the initiator information to the preset approval client so that the preset approval client approves the task initiation request based on the user information corresponding to the target task type; where there are multiple preset approval clients;

[0080] S113. When all the preset approval clients approve the task initiation request, allow the initiator to execute the target task; and execute the step of acquiring the construction site image collected by the head-mounted device as the target image to be recognized during the execution of the target task;

[0081] S114. If there is a preset approval client that does not approve the task initiation request, reject the initiator from executing the target task.

[0082] Construction site personnel can send a task initiation request to the server through a head-mounted device. The task initiation request can include the requested target task type and initiator information. The above task types can include safety inspection tasks, daily inspection tasks, weekly inspection tasks, special inspection tasks, dangerous operation applications, safety acceptance, and electrical box inspection tasks, etc. The server can determine whether the initiator has the target task permission based on the corresponding relationship between the preset service type and the user permission. If so, the target task is allowed to be executed; if not, the execution of the target task is rejected.

[0083] As a possible implementation, different approval clients can be preset for different tasks. There can be multiple approval clients corresponding to one task. Only when the sponsor information is approved by all the approval clients corresponding to the task can it be determined that the sponsor has the execution authority for the task, that is, the sponsor is allowed to execute the task. Otherwise, the sponsor is not allowed to execute the task.

[0084] If the sponsor is allowed to execute the target task, the inspection checklist corresponding to the target task can be issued. The inspection checklist can include the items required to be inspected in the safety specifications corresponding to the target task, so that the construction site personnel can collect construction site images based on the inspection checklist, and then obtain the corresponding construction site inspection results through the scene recognition model and the safety recognition model.

[0085] In a possible embodiment, the above target task can include multiple processes. Exemplarily, the fire operation process can include permission application, fire operation, and post-operation processing. The processes included in each task can be set according to the actual application scenario. At the same time, safety acceptance criteria can be set for each process. During the process of relevant personnel executing the process, relevant documents, images, etc. can be uploaded to the server, so that the server can track the compliance of each process, thereby improving the execution safety of the target task.

[0086] In practical applications, there are tasks that need to be performed periodically, such as inspection tasks, weekly inspection tasks, etc. For these tasks, the server can issue the tasks to the preset client according to the preset inspection cycle to remind the relevant personnel to execute the periodic task, and can also perform compliance detection on the inspection results of the periodic task, thereby improving the task inspection efficiency and the safety of the construction site.

[0087] In a possible embodiment, the above server can also access an inference large model. The inference large model can implement question answering. Specifically, audio can be collected through a head-mounted device, and the question information contained in the audio can be extracted and output to the inference large model to obtain the answer data answered by the inference large model for the question. Returning the answer data to the head-mounted device for display can achieve instant question answering and improve construction efficiency. In practical applications, through natural language interaction, scenario questions such as "high-bay formwork acceptance standards" and "temporary power supply configuration requirements" can be answered in real time, and the response accuracy rate reaches 95%. Again, the inspection checklist corresponding to the safety inspection task can also be generated through the inference model. Specifically, the inference large model can automatically output a customized inspection checklist based on the construction progress, synchronously associate standard articles with typical cases, and improve the inspection pertinence.

[0088] In a possible implementation, the server can also integrate other collaboration tools, such as shared documents, real-time chat, video conferencing, etc., to make remote collaboration more efficient and convenient.

[0089] The construction site management method of the integrated head-mounted device provided by the present invention can be implemented through software, which can include a server and a front-end display page. Among them, the server can execute the above S101-S104, and the front-end display page is used to display information, such as the detection result, approval process, process compliance, etc.

[0090] As Figure 2 shown, Figure 2 FIG. is a schematic flowchart of a construction site management method of an integrated head-mounted device provided by an embodiment of the present invention. This construction site management method can realize hidden danger investigation and treatment, dangerous operation, safety acceptance, and electrical box inspection. Among them, the hidden danger investigation and treatment includes that relevant personnel initiate a safety inspection task to conduct a safety inspection on the construction site according to a preset list. If a daily inspection task is initiated, a daily inspection of the construction site is carried out according to the daily inspection items. If a weekly / special inspection task is initiated, a gathering notice can be sent to relevant personnel to enable them to conduct on-site inspections. If a safety hazard is detected at the construction site, the hidden danger can be treated according to the preset measures, and the whole process of hidden danger rectification is monitored. After the hidden danger rectification is completed, the corresponding hidden danger can be re-audited to determine whether the safety hazard has been eliminated after rectification.

[0091] In the dangerous operation scenario, dangerous operation approval and dangerous operation detection and monitoring can be carried out. Specifically, each process of the dangerous operation can be monitored to determine whether it complies with safety specifications, and the operation result can be audited after the operation is completed.

[0092] In the safety acceptance scenario, after receiving a safety acceptance task initiation request, construction site images can be collected based on a preset safety acceptance list, and each safety recognition model can be used to conduct a safety acceptance of the construction site. If an abnormal device, that is, a problem device, appears, problem device handling information can be pushed to the relevant client to enable relevant personnel to handle the problem device.

[0093] The above management software can also automatically generate inspection tasks. For example, the electrical box can be inspected. Specifically, electrical box inspection can be carried out, including inspection of the appearance, identification, internal electrical components, and wiring of the electrical box. In the case of an abnormal detection result, the detection situation can be pushed to the relevant client to enable the corresponding personnel to repair the electrical box.

[0094] In the above various operation scenarios, corresponding task ledgers, hidden danger ledgers, etc. can be generated after the operation. The task ledger is used to record the task. For example, it can record information such as the task initiator information, execution time, task type, approval process, on-site images, detection results, maintenance records, etc. The hidden danger ledger can include information such as the detected hidden danger type, inspection time, hidden danger rectification situation, person in charge, etc., so as to facilitate the query of task historical records.

[0095] Applying the embodiments of the present invention, by combining a software platform and wearable devices, all parties involved in the construction site, including workers, project managers, engineers, and safety experts, can share information and resources on a unified interface. Especially when dealing with complex construction tasks, remote experts can provide detailed technical support and solutions through the platform. For parts of the construction site that are difficult to inspect or require collaborative inspection, remote video acceptance is achieved. During this process, project experts provide assistance to on-site technicians through remote video, sharing images and standard acceptance results in real time. The wearable device is built-in with standard inspection processes and relevant specifications, providing detailed operation guidance and practical training for on-site engineers. With the support of AI / AR technology, this solution significantly improves the safety guarantee level of the project and at least increases the safety inspection efficiency by 10%.

[0096] In addition, this solution automatically identifies potential safety hazards on site through AI algorithms, significantly improving the efficiency of construction safety management. Using advanced computer vision technology, the images captured by on-site cameras are analyzed in real time to automatically identify various inspection scenarios, such as fixed-point inspections, hot work operations, confined space operations, etc., and accurately identify common safety hazards such as the compliance of electrical box wiring and electrical box accessories. Compared with traditional manual inspections, the automatic identification system not only greatly improves the speed and accuracy of hazard identification, but also significantly reduces the risk of human negligence, ensuring that safety hazards can be discovered and handled in a timely manner, effectively reducing the incidence of safety accidents, and providing a solid guarantee for the safety of workers and the site.

[0097] Based on the same inventive concept, the embodiments of the present invention also provide a construction site safety management system integrated with a head-mounted device, as Figure 3 shown. The system 300 may include a head-mounted device 301 and a server 302;

[0098] Among them, the head-mounted device 301 is used to collect images of the construction site and send the images of the construction site to the server;

[0099] The server 302 is configured to input the construction site image as a target image into a preset scene recognition model, and obtain a scene recognition result output by the preset scene recognition model based on the target image as a target scene, where the target scene at least includes fixed-point inspection, high-altitude operation, hot work operation, or confined space operation;

[0100] Based on a safety recognition strategy preset for the target scene, the target image is recognized to obtain a safety detection result of the target image, where the safety recognition strategy includes one or more of safety helmet recognition, electrical box recognition, hot work operation recognition, reflective vest recognition, high-altitude operation safety belt recognition, and edge protection detection;

[0101] Send the safety detection result to a preset client so that the preset client displays the safety detection result.

[0102] In a possible embodiment, a safety helmet recognition model and a reflective vest recognition model are integrated in the head-mounted device; the head-mounted device is configured to input the construction site image into the safety helmet recognition model to identify whether the safety helmet is worn in compliance in the construction site image, and / or input the construction site image into the reflective vest recognition model to identify whether the reflective vest is worn in compliance in the construction site image.

[0103] In a possible embodiment, the server is configured to determine a risk level corresponding to the safety detection result according to a preset risk assessment standard when the safety detection result does not meet the safety specification; where the preset risk assessment standard includes risk severity, occurrence probability, and impact range;

[0104] Send the risk level of the safety detection result to the preset client so that the preset client displays the risk level;

[0105] Before obtaining the construction site image collected by the head-mounted device as a target image to be recognized, the method further includes:

[0106] Receive a task initiation request, where the task initiation request includes a target task type and initiator information, and the target task type includes safety inspection, daily inspection, dangerous operation application, and safety acceptance;

[0107] Send the initiator information to a preset approval client so that the preset approval client approves the task initiation request based on the user information corresponding to the target task type; where there are multiple preset approval clients;

[0108] In the case where all the preset approval clients approve the task initiation request, allowing the initiator to execute the target task; and in the process of executing the target task, executing the step of acquiring the construction site image collected by the head-mounted device as the target image to be identified;

[0109] If there is a preset approval client that has not approved the task initiation request, the initiator is denied execution of the target task;

[0110] The server includes a large deduction model, and the head-mounted device is also used to collect audio from the construction site and send the audio to the server;

[0111] The server is used to extract the question information contained in the audio and output the question information to the deduction large model;

[0112] Obtaining the question and answer results output by the deduction large model and returning them to the head mounted device;

[0113] The server is also used to regularly send inspection tasks to the client with the inspection task authority according to the preset inspection cycle, and at the same time send the inspection list included in the inspection task.

[0114] Among them, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved in the present invention are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0115] The exemplary embodiment of the present invention further provides an electronic device, comprising: at least one processor; and a memory connected to the at least one processor in communication. The memory stores a computer program that can be executed by the at least one processor, and the computer program is used to enable the electronic device to perform a method according to an embodiment of the present invention when executed by the at least one processor.

[0116] Exemplary embodiments of the present invention also provide a non-transitory computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor of a computer, is used to cause the computer to perform a method according to an embodiment of the present invention.

[0117] An exemplary embodiment of the present invention further provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor of a computer, the computer is used to enable the computer to perform a method according to an embodiment of the present invention.

[0118] refer to Figure 4, a block diagram of an electronic device 400 that can be a server or a client of the present invention will now be described. It is an example of a hardware device that can be applied to various aspects of the present invention. The electronic device is intended to represent various forms of digital electronic computer devices, such as, laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, personal digital processors, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0119] As Figure 4 shown, the electronic device 400 includes a computing unit 401, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 402 or a computer program loaded from a storage unit 408 into a random access memory (RAM) 403. In the RAM 403, various programs and data required for the operation of the electronic device 400 can also be stored. The computing unit 401, the ROM 402, and the RAM 403 are connected to each other via a bus 404. An input / output (I / O) interface 405 is also connected to the bus 404.

[0120] Multiple components in the electronic device 400 are connected to the I / O interface 405, including: an input unit 406, an output unit 407, a storage unit 408, and a communication unit 409. The input unit 406 can be any type of device that can input information into the electronic device 400. The input unit 406 can receive input digital or character information, and generate key signal inputs related to the user settings and / or function controls of the electronic device. The output unit 407 can be any type of device that can present information, and can include but is not limited to a display, a speaker, a video / audio output terminal, a vibrator, and / or a printer. The storage unit 408 can include but is not limited to a magnetic disk, an optical disk. The communication unit 409 allows the electronic device 400 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks, and can include but is not limited to a modem, a network card, an infrared communication device, a wireless communication transceiver, and / or a chipset, such as a Bluetooth™ device, a WiFi device, a WiMax device, a cellular communication device, and / or the like.

[0121] The computing unit 401 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 401 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 401 executes the various methods and processes described above. For example, in some embodiments, the construction site safety management method of the integrated head-mounted device described above can be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as the storage unit 408. In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device 400 via the ROM 402 and / or the communication unit 409. In some embodiments, the computing unit 401 can be configured to execute the construction site safety management method of the integrated head-mounted device described above in any other suitable manner (e.g., by means of firmware).

[0122] The program code for implementing the method of the present invention can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing devices, such that when the program codes are executed by the processor or controller, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The program codes can be executed entirely on the machine, partially on the machine, executed partially on the machine and partially on a remote machine as an independent software package, or executed entirely on a remote machine or server.

[0123] In the context of the present invention, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0124] As used in this invention, the terms "machine-readable medium" and "computer-readable medium" refer to any computer program product, apparatus, and / or device (e.g., a magnetic disk, an optical disk, a memory, a programmable logic device (PLD)) used to provide machine instructions and / or data to a programmable processor, including a machine-readable medium that receives machine instructions as a machine-readable signal. The term "machine-readable signal" refers to any signal used to provide machine instructions and / or data to a programmable processor.

[0125] In order to provide for interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can also be used to provide for interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, speech input, or tactile input).

[0126] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0127] A computer system can include a client and a server. The client and server are generally remote from each other and typically interact through a communication network. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship to each other.

Claims

1. A construction site safety management method integrating a head-mounted device, characterized in that, Applied to the server side, the method includes: Obtain the construction site image collected by the head-mounted device as the target image to be recognized; Input the target image into a preset scene recognition model, and obtain the scene recognition result output by the preset scene recognition model based on the target image as the target scene, where the target scene at least includes fixed-point inspection, high-altitude operation, hot work operation, or confined space operation; Based on the safety recognition strategy pre-set for the target scene, recognize the target image to obtain the safety detection result of the target image, where the safety recognition strategy includes one or more of safety helmet recognition, electrical box recognition, hot work operation recognition, reflective vest recognition, high-altitude operation safety belt recognition, and edge protection detection; Send the safety detection result to a preset client so that the preset client displays the safety detection result.

2. The method according to claim 1, wherein The method further includes: In the case where the safety detection result does not meet the safety specifications, determine the risk level corresponding to the safety detection result according to the preset risk assessment criteria; where the preset risk assessment criteria include risk severity, occurrence probability, and impact range; Send the risk level of the safety detection result to the preset client so that the preset client displays the risk level.

3. The method according to claim 1, characterized in that Before obtaining the construction site image collected by the head-mounted device as the target image to be recognized, the method further includes: Receive a task initiation request, where the task initiation request includes a target task type and initiator information, and the target task type includes safety inspection, daily inspection, dangerous operation application, and safety acceptance; Send the initiator information to a preset approval client so that the preset approval client approves the task initiation request based on the user information corresponding to the target task type; where there are multiple preset approval clients; In the case where each of the preset approval clients approves the task initiation request, allow the initiator to execute the target task; and execute the step of obtaining the construction site image collected by the head-mounted device as the target image to be recognized during the execution of the target task; If there is a preset approval client that does not approve the task initiation request, reject the initiator from executing the target task.

4. The method according to claim 1, wherein The server side contains a deduction large model, and the method further includes: Obtain the audio collected by the head-mounted device; Extract the problem information contained in the audio and output the problem information to the deduction large model; Obtain the Q&A result output by the deduction large model and return it to the head-mounted device.

5. The method according to claim 1, wherein The method further includes: Regularly issue inspection tasks to the clients with the inspection task permissions according to a preset inspection cycle, and at the same time issue the inspection list included in the inspection tasks.

6. A construction site safety management system integrated with a head-mounted device, characterized in that, The system includes a head-mounted device and a server side, where the head-mounted device is used to collect construction site images and send the construction site images to the server side; The server is used to input the construction site image as a target image into a preset scene recognition model, and obtain the scene recognition result output by the preset scene recognition model based on the target image as the target scene, where the target scene at least includes fixed-point inspection, high-altitude operation, hot work operation or confined space operation; Based on the safety recognition strategy preset for the target scene, the target image is recognized to obtain the safety detection result of the target image, where the safety recognition strategy includes one or more of safety helmet recognition, electrical box recognition, hot work operation recognition, reflective vest recognition, high-altitude operation safety belt recognition, and edge protection detection; Send the safety detection result to a preset client so that the preset client can display the safety detection result.

7. The system according to claim 6, wherein The helmet-mounted device integrates a safety helmet recognition model and a reflective vest recognition model; the helmet-mounted device is used to input the construction site image into the safety helmet recognition model to identify whether the safety helmet is worn in compliance in the construction site image, and / or input the construction site image into the reflective vest recognition model to identify whether the reflective vest is worn in compliance in the construction site image.

8. The system according to claim 6, wherein The server is used to determine the risk level corresponding to the safety detection result according to the preset risk assessment criteria when the safety detection result does not meet the safety specifications; where the preset risk assessment criteria include risk severity, occurrence probability, and impact range; Send the risk level of the safety detection result to the preset client so that the preset client can display the risk level; Before obtaining the construction site image collected by the helmet-mounted device as the target image to be recognized, the method further includes: Receiving a task initiation request, where the task initiation request includes a target task type and initiator information, and the target task type includes safety inspection, daily inspection, dangerous operation application, and safety acceptance; Send the initiator information to a preset approval client so that the preset approval client can approve the task initiation request based on the user information corresponding to the target task type; where there are multiple preset approval clients; When all the preset approval clients approve the task initiation request, allow the initiator to execute the target task; and execute the step of obtaining the construction site image collected by the helmet-mounted device as the target image to be recognized during the execution of the target task; If there is a preset approval client that does not approve the task initiation request, reject the initiator from executing the target task; The server contains a deduction large model, and the helmet-mounted device is also used to collect construction site audio and send the audio to the server; The server is used to extract the problem information contained in the audio and output the problem information to the deduction large model; Obtain the Q&A result output by the deduction large model and return it to the helmet-mounted device; The server is further configured to regularly send inspection tasks to the client with the permission of the inspection tasks according to a preset inspection cycle, and simultaneously send the inspection list included in the inspection tasks.

9. An electronic device, comprising: a processor; and a memory storing a program, wherein the program includes instructions that, when executed by the processor, cause the processor to execute the method according to any one of claims 1-5.

10. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are used to cause a computer to execute the method according to any one of claims 1-5.

Images