Pipeline construction potential safety hazard automatic closed-loop management system based on AI video

Abstract

The invention provides an AI video-based pipeline construction potential safety hazard automatic closed-loop management system and method, and the system comprises an information collector which is used for collecting the video information of a construction site; the AI server is used for identifying violation behaviors in the video information and generating an alarm signal; and the alarm terminal is used for recording a violation correction process according to the alarm signal, obtaining violation correction process information and uploading the violation correction process information to the AI server. According to the AI video-based pipeline construction potential safety hazard automatic closed-loop management system and method provided by the embodiment of the invention, the AI video identification technology is utilized to identify the violation behaviors in the construction, the automation and intelligence of the construction safety management are realized, the violation correction process is recorded through the alarm terminal, and the management efficiency is improved. According to the invention, real-time automatic identification and alarm of potential safety hazards and recording and uploading of violation correction processes are realized, and closed-loop management of safety events is realized, so that the safety accident risk is reduced, and the construction safety management efficiency is improved.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence technology, and specifically to an automated closed-loop management system for pipeline construction safety hazards based on AI video, an automated closed-loop management method for pipeline construction safety hazards based on AI video, an electronic device, and a computer storage medium. Background Art

[0002] The situation at the pipeline construction site is complex, safety hazards are everywhere, the security situation is becoming increasingly severe, and it is urgent to improve the management and control methods. At present, the more common technical prevention and control means used in pipeline construction site safety management is to use a video surveillance system to transmit the video of the construction site back to the background in real time, and a dedicated person will find non-conformities from the video. This method has played a certain positive role in discovering non-conformities and strengthening safety supervision, but it still has shortcomings such as equipment limitations, high labor consumption, low prevention and control efficiency, and non-real-time prevention and control. Therefore, pipeline construction safety management urgently needs to strengthen the automation and intelligent management of construction safety through advanced technical means to fill the gap in the intelligent management of construction safety during the pipeline construction period. In addition, in the construction safety monitoring method based on video image recognition in the existing technology, a large number of training samples need to be collected. It does not have a good certainty in the judgment of violations and the handling of violation alarms, and there is a great possibility of misjudgment, and the cost of manual review will increase to a great extent. Summary of the invention

[0003] The present disclosure aims to solve at least one of the technical problems existing in the prior art, and provides an automated closed-loop management system for pipeline construction safety hazards based on AI video, an automated closed-loop management method for pipeline construction safety hazards based on AI video, an electronic device and a computer storage medium.

[0004] One aspect of the present disclosure provides an automated closed-loop management system for pipeline construction safety hazards based on AI video, including:

[0005] Information collector, used to collect video information of the construction site;

[0006] An AI server, used to identify illegal behaviors in the video information and generate an alarm signal;

[0007] The alarm terminal is used to record the violation correction process according to the alarm signal, obtain the violation correction process information, and upload the violation correction process information to the AI ​​server.

[0008] Optionally, the AI ​​server includes a video recognition module, a violation judgment module and a violation alarm module:

[0009] The video recognition module is used to identify possible violations in the video information;

[0010] The violation judgment module is used to judge whether the possible violation behavior is a violation behavior, and send the violation behavior to the violation alarm module;

[0011] The violation alarm module is used to generate a violation record and an alarm signal according to the violation behavior.

[0012] Optionally, the video recognition module includes:

[0013] A multi-type single model recognition layer, used to identify possible violation types in the video information;

[0014] The single-type single-model recognition layer is used to call the corresponding recognition model according to the violation type to identify possible violation behaviors.

[0015] Optionally, the traffic violation judgment module is used to judge whether the possible traffic violation is a traffic violation based on the number of times the possible traffic violation occurs within a preset time range.

[0016] Optionally, the violation judgment module is used to judge whether multiple violation behaviors are of the same type within the rectification time range. If they are of the same type, the violation judgment module only sends the first violation behavior of the same type to the violation alarm module.

[0017] Optionally, it also includes: a communication module, used to realize communication between the AI ​​server and the alarm terminal.

[0018] Another aspect of the present disclosure further provides an automated closed-loop management method for pipeline construction safety hazards based on AI video, including:

[0019] Collect video information from the construction site;

[0020] Identifying illegal behaviors in the video information and generating an alarm signal;

[0021] The violation correction process is recorded according to the alarm signal to obtain violation correction process information, and the violation correction process information is uploaded.

[0022] Optionally, the identifying the illegal behavior in the video information and generating an alarm signal includes:

[0023] Identifying possible traffic violations in the video information;

[0024] Determining whether the possible violation is a violation;

[0025] A violation record and an alarm signal are generated according to the violation behavior.

[0026] Another aspect of the present disclosure further provides an electronic device, characterized in that it includes:

[0027] one or more processors;

[0028] A storage unit is used to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors can implement the automated closed-loop management method for pipeline construction safety hazards based on AI video as described above.

[0029] Another aspect of the present disclosure provides a computer storage medium, characterized in that computer-readable instructions are stored therein, and when the computer-readable instructions are executed by a processor, the method for automated closed-loop management of safety hazards in pipeline construction based on AI video as described above is implemented.

[0030] The AI ​​video-based automated closed-loop management system and method for pipeline construction safety hazards of the disclosed embodiments utilize AI video recognition technology to identify violations during construction, thereby realizing automation and intelligence in construction safety management, reducing or eliminating the occurrence of construction safety accidents, and recording the violation correction process through an alarm terminal, thereby realizing real-time automatic identification of safety hazards, alarming, and recording and uploading of the violation correction process, and realizing closed-loop management of safety incidents, thereby reducing the risk of safety accidents, improving the efficiency of construction safety management, reducing the labor intensity of supervisors, and promoting the progress of intelligent management of pipeline construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 A schematic block diagram of an example electronic device for implementing an automated closed-loop management system and method for safety hazards in pipeline construction based on AI video according to an embodiment of the present disclosure;

[0032] Figure 2 This is a structural schematic diagram of an automated closed-loop management system for pipeline construction safety hazards based on AI video according to another embodiment of the present disclosure;

[0033] Figure 3 This is a site layout diagram of an automated closed-loop management system for pipeline construction safety hazards based on AI video according to another embodiment of the present disclosure;

[0034] Figure 4 This is a flow chart of an automated closed-loop management method for safety hazards in pipeline construction based on AI video according to another embodiment of the present disclosure;

[0035] Figure 5 This is a flow chart of an automated closed-loop management method for safety hazards in pipeline construction based on AI video according to another embodiment of the present invention. DETAILED DESCRIPTION

[0036] In order to enable those skilled in the art to better understand the technical solution of the present disclosure, the present disclosure is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0037] In order to enable those skilled in the art to better understand the technical solution of the present disclosure, the present disclosure is further described in detail below in conjunction with the accompanying drawings and specific implementation methods. Obviously, the described embodiments are part of the embodiments of the present disclosure, rather than all of the embodiments. Based on the described embodiments of the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure.

[0038] Unless otherwise specified, the technical terms or scientific terms used in the present disclosure shall have the usual meanings understood by persons with ordinary skills in the field to which the present disclosure belongs. The terms "including" or "comprising" used in the present disclosure do not limit the shapes, numbers, steps, actions, operations, components, originals and / or groups thereof mentioned, nor do they exclude the presence or addition of one or more other different shapes, numbers, steps, actions, operations, components, originals and / or groups thereof, or the addition of these. In the description of the present disclosure, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0039] It should be mentioned before discussing in more detail that some exemplary embodiments are described as processes or methods depicted as flow charts. Although the flow charts describe the operations (steps) as sequential processes, many of the operations therein can be implemented in parallel, concurrently or simultaneously. In addition, the order of the operations can be rearranged. The process can be terminated when the operation is completed, but can also have additional steps not included in the accompanying drawings. The process can correspond to a method, function, procedure, subroutine, subprogram, etc.

[0040] First, refer to Figure 1 To describe an example electronic device for implementing an AI video-based automated closed-loop management system and method for safety hazards in pipeline construction according to an embodiment of the present disclosure.

[0041] like Figure 1 As shown, the electronic device 300 includes one or more processors 310, one or more storage devices 320, an input device 330, an output device 340, etc. These components are interconnected through a bus system and / or other forms of connection mechanisms 350. It should be noted that Figure 1 The components and structures of the electronic device shown are merely exemplary and non-limiting. The electronic device may also have other components and structures as required.

[0042] The processor 310 may be a central processing unit (CPU) or other forms of processing units having data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device to perform desired functions.

[0043] The storage device 320 may include one or more computer program products, and the computer program product may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory (cache), etc. The non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor may run the program instructions to implement the client functions (implemented by the processor) and / or other desired functions in the embodiments of the present disclosure described below. Various applications and various data may also be stored in the computer-readable storage medium, for example, various data used and / or generated by the application, etc.

[0044] The input device 330 may be a device used by a user to input instructions, and may include one or more of a keyboard, a mouse, a microphone, a touch screen, and the like.

[0045] The output device 340 may output various information (eg, images or sounds) to the outside (eg, a user), and may include one or more of a display, a speaker, and the like.

[0046] It should be noted that the AI ​​video-based automated closed-loop management system and method for pipeline construction safety hazards in this embodiment can identify the following violations: ① Failure to correctly wear work clothes, work shoes, safety helmets and other personal protective equipment as required; ② Walking on steel pipes or standing on steel pipes to direct operations; ③ Not using a slip rope to directly hold and drag steel pipes during pipe laying, loading and unloading operations; ④ The anti-detachment baffle of the hook is missing or its performance fails; ⑤ Failure to provide safety reminders for outsiders to enter the construction site or outsiders entering the construction site at will without being stopped in time; ⑥ Using excavator bucket teeth for lifting and hoisting operations. In addition to this, technical personnel in this field can also use the closed-loop management system and method in this embodiment to automatically identify other non-conformities related to construction safety according to actual usage needs, and no specific restrictions are made in this embodiment.

[0047] Next, we will refer to Figure 2 and Figure 3 An automated closed-loop management system for safety hazards in pipeline construction based on AI video is described according to another embodiment of the present disclosure.

[0048] like Figure 2 and Figure 3 As shown, the automated closed-loop management system for pipeline construction safety hazards based on AI video includes an information collector, an AI server and an alarm terminal.

[0049] Specifically, the information collector is used to collect video information of the construction site and send the video information to the AI ​​server. Exemplarily, the information collector can be a camera, and those skilled in the art can select the information collector according to actual use requirements, and no specific limitation is made in this embodiment.

[0050] Specifically, the AI ​​server is used to receive the video information sent by the information collector and identify the illegal behavior in the video information, thereby generating a corresponding alarm signal according to the illegal behavior, and then sending the alarm signal to the alarm terminal to activate the recording function of the alarm terminal. Exemplarily, the AI ​​server is constructed with small-sized, low-power ARM and NPU, and the video recognition module, violation judgment module, and violation alarm module are deployed in the AI ​​server, and it has built-in 4G and WiFi communication modules to support remote and LAN communications.

[0051] Specifically, the alarm terminal is used to receive an alarm signal, and according to the alarm signal, activate the recording function of the alarm terminal, record the violation correction process, obtain the violation correction process information, and upload the violation correction process information to the AI ​​server. Exemplarily, the alarm terminal adopts a device modified from a law enforcement recorder, which has the functions of receiving alarm signals, viewing alarm information, and recording the violation correction process. Exemplarily, when a violation is found, the AI ​​server sends the alarm signal to the alarm terminal in the form of a text message, and simultaneously activates the alarm light and recording function of the alarm terminal, records the violation correction process and transmits it to the AI ​​server in real time. Furthermore, if a remote server is configured, the violation correction process information can also be transmitted back to the remote server through the AI ​​server.

[0052] The AI ​​video-based automated closed-loop management system for pipeline construction safety hazards of the disclosed embodiment utilizes AI video recognition technology to identify violations during construction, realizes automation and intelligence in construction safety management, reduces or eliminates the occurrence of construction safety accidents, and records the process of correcting violations through the alarm terminal, thereby realizing real-time automatic identification of safety hazards, alarming, and recording and uploading the process of correcting violations, and realizing closed-loop management of safety incidents, thereby reducing the risk of safety accidents, improving the efficiency of construction safety management, reducing the labor intensity of supervisors, and promoting the process of intelligent management of pipeline construction. In addition, the information collector, AI server, and alarm terminal jointly realize front-end recognition, which can better adapt to the complex environment of the construction site. Compared with the back-end recognition solution, it has the advantages of not relying on the communication conditions on site, real-time alarm, and more room for promotion.

[0053] Next, combine Figure 2 A specific example illustrates another embodiment of the present disclosure, which is an automated closed-loop management system for pipeline construction safety hazards based on AI video.

[0054] For example, Figure 2 As shown, the AI ​​server includes a video recognition module, a violation judgment module and a violation alarm module.

[0055] Exemplarily, the video recognition module is used to receive video information sent by the information collector, and identify the video information to identify possible violations in the video information, and then send the possible violations to the violation judgment module.

[0056] Exemplarily, the violation judgment module is used to judge whether the possible violation is a violation, realize the screening of the video recognition result, and finally determine whether the recognition result is a violation, and then send the violation to the violation alarm module.

[0057] Exemplarily, the traffic violation alarm module is used to receive the traffic violation behavior sent by the traffic violation judgment module, and generate a traffic violation record and a corresponding alarm signal according to the traffic violation behavior, and then send the alarm signal to the alarm terminal for alarm.

[0058] The AI ​​video-based automated closed-loop management system for pipeline construction safety hazards of the disclosed embodiment recognizes and identifies illegal behaviors through a video recognition module, a violation judgment module, and a violation alarm module, thereby improving recognition accuracy.

[0059] The following is a specific example of a video recognition module according to another embodiment of the present disclosure.

[0060] Exemplarily, the video recognition module adopts two-layer recognition, specifically including a first-layer multi-type single-model recognition layer and a second-layer single-type single-model recognition layer.

[0061] Exemplarily, the multi-type single model recognition layer is used to identify possible violation types in the video information. In the multi-type single model recognition layer, a pre-trained multi-type single model is used for recognition. The multi-type single model refers to a model that uses a single model to identify multiple types of violations. This model cannot provide a roughly equal number of positive and negative samples during training, and therefore must be based on a large amount of training material to achieve satisfactory training results.

[0062] Exemplarily, a single-type single-model recognition layer is used to call the corresponding recognition model according to the violation type to identify possible violations. In the single-type single-model recognition layer, a pre-trained single-type single model is used. A single-type single model refers to a model that only identifies one type of violation. This model can provide an equal number of positive and negative samples for each violation type, and can achieve satisfactory training results with a small amount of material. However, for applications that identify multiple violations, this model needs to run multiple models at the same time, which will increase the load on the AI ​​server.

[0063] In actual work, in order to improve the accuracy of recognition, for each violation recognition type in the multi-type single model, the requirement for the amount of training materials is very large, which makes it difficult to collect training materials. Training can only be carried out on the basis of a limited amount of materials, resulting in a low accuracy rate of video recognition; and under the premise of the same amount of training materials, the recognition accuracy of a single-type single model is much higher. If a single-type single-model recognition method is directly adopted, for applications that identify multiple types of violations, it means that multiple models must be run at the same time, resulting in an increased load on the AI ​​server and making the AI ​​server unstable. In this embodiment, a two-layer design is adopted, and a multi-type single model can be used to identify possible violation types, and then the corresponding single-type single model is called for confirmation according to the violation type. This reduces the load on the AI ​​server and improves the accuracy of recognition.

[0064] The AI ​​video-based automated closed-loop management system for pipeline construction safety hazards of the disclosed embodiment improves the recognition accuracy on the basis of less training materials through the multi-type single model recognition layer and the single-type single model recognition layer. The two-layer recognition method adopted in the present disclosure can minimize the training cost and cycle of the recognition model, as well as the load of the AI ​​server. Because the present disclosure targets multiple violation units and requires a large amount of materials, if the multi-type single model method is still used at this time, it will not only increase the cost and cycle of model training, but also make it difficult to obtain the ideal recognition accuracy. The present disclosure reduces the cost and cycle of model training and improves the recognition accuracy.

[0065] The following is a specific example of a traffic violation judgment module according to another embodiment of the present disclosure.

[0066] Exemplarily, the violation judgment module counts the possible violation behaviors received within a preset time range, obtains the number of occurrences, and then judges whether the possible violation behaviors are violation behaviors based on the number of occurrences. Specifically, in practical applications, false positives of video recognition generally appear in a jumpy manner rather than continuously, so the design uses the number of occurrences of recognition results within a specified time range to filter out false positive results, thereby improving the accuracy of recognition.

[0067] For example, the relationship between the number of times a possible violation occurs and the preset false alarm threshold can be compared. If the number of times a possible violation occurs is greater than the false alarm threshold, it can be determined that it does not appear in a jumpy manner, and it is determined to be a violation. For example, if the false alarm threshold is set to 4 times, then the possible violation occurs 4 times or more, and it can be determined to be a violation. Those skilled in the art can preset the time range and false alarm threshold according to actual usage, and no specific limitation is made in this embodiment.

[0068] Exemplarily, the violation judgment module can be used to judge whether multiple violations are of the same type within the rectification time range. If they are of the same type, the violation judgment module only sends the first violation of the same type to the violation alarm module. Specifically, the frequency of video recognition is 5-15 frames per second, that is, for the same violation, 5-15 alarms are triggered per second. Such frequent alarms give a very poor user experience. Therefore, a given rectification time is adopted in the design. Within the rectification time range, the same violation will no longer be alarmed. The so-called rectification time is the time to correct the violation after the violation occurs. This ensures that the same violation is only alarmed once, which improves the user experience and is more in line with the actual situation on site.

[0069] For example, in a specific use process, after the violation judgment module confirms the first violation, it compares the subsequent violation with the first violation within the rectification time range. If the types are the same, the subsequent violation is not sent to the violation alarm module, that is, no alarm is issued. For example, the rectification time range can be half an hour. Those skilled in the art can preset the rectification time range according to actual usage, and no specific limitation is made in this embodiment.

[0070] The AI ​​video-based automated closed-loop management system for pipeline construction safety hazards in the disclosed embodiment reduces the false alarm rate of video recognition by judging whether a possible violation is a violation by the number of times it occurs within a preset time range, and avoids frequent system alarms by judging whether multiple violations are of the same type within the rectification time range.

[0071] Below, a specific example is given to illustrate an automated closed-loop management system for pipeline construction safety hazards based on AI video according to another embodiment of the present disclosure.

[0072] Exemplarily, the AI ​​video-based automated closed-loop management system for pipeline construction safety hazards also includes a remote server for implementing remote service functions.

[0073] Exemplarily, the AI ​​video-based automated closed-loop management system for safety hazards in pipeline construction also includes a communication module for realizing communication between the AI ​​server and the alarm terminal, specifically, for realizing communication between the violation alarm module and the alarm terminal, for example, sending the alarm signal generated by the violation alarm module to the alarm terminal, or uploading the violation correction process information recorded by the alarm terminal to the violation alarm module in the AI ​​server.

[0074] Exemplarily, the communication module is also used to realize the communication between the AI ​​server and the remote server, and between the alarm terminal and the remote server, for example, sending the violation correction process information from the alarm terminal to the remote server, or sending the violation correction process information from the AI ​​server to the remote server. Exemplarily, the communication module can also realize the communication between other components in the automated closed-loop management system, and the communication between the automated closed-loop management system and other external devices, which is not specifically limited in this embodiment. Exemplarily, the communication module can use wireless communication methods such as Bluetooth and WiFi.

[0075] Specifically, during use, the camera sends the captured scene to the video recognition module, the video recognition module recognizes the video and sends the recognition result to the violation judgment module, the violation judgment module determines whether it is a violation based on the recognition result, if it is a violation, the violation alarm module is called to generate a violation record and alarm signal, and store the video of the violation, and at the same time send the alarm signal to the alarm terminal and the remote server through the communication module.

[0076] Next, combine Figure 4 Another embodiment of the present disclosure is described as an automated closed-loop management method for pipeline construction safety hazards based on AI video, including:

[0077] S100: Collect video information of the construction site.

[0078] S200: Identify illegal behaviors in the video information and generate an alarm signal.

[0079] S300: Recording the violation correction process according to the alarm signal, obtaining violation correction process information, and uploading the violation correction process information.

[0080] For example, Figure 5 As shown, step S200 includes:

[0081] S210: Identify possible traffic violations in the video information.

[0082] S220: Determine whether the possible violation is a violation.

[0083] S230: Generate a violation record and an alarm signal according to the violation behavior.

[0084] The AI ​​video-based automated closed-loop management method for pipeline construction safety hazards of the disclosed embodiment utilizes AI video recognition technology to identify violations during construction, realizes automation and intelligence in construction safety management, reduces or eliminates the occurrence of construction safety accidents, records the process of correcting violations, realizes real-time automatic identification and alarm of safety hazards, and records and uploads the process of correcting violations, realizes closed-loop management of safety incidents, thereby reducing the risk of safety accidents, improving the efficiency of construction safety management, reducing the labor intensity of supervisors, and promoting the process of intelligent management of pipeline construction.

[0085] Furthermore, this embodiment also discloses an electronic device, including:

[0086] one or more processors;

[0087] A storage unit is used to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors can implement the automated closed-loop management method for pipeline construction safety hazards based on AI video as described above.

[0088] Furthermore, the present embodiment also discloses a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the automated closed-loop management method for pipeline construction safety hazards based on AI video as described above can be implemented.

[0089] The computer-readable medium may be included in the apparatus, device, or system of the present disclosure, or may exist independently.

[0090] Among them, computer-readable storage media can be any tangible media that contains or stores a program, which can be an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment. More specific examples include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, an optical fiber, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0091] The computer-readable storage medium may also include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code, specific examples of which include but are not limited to electromagnetic signals, optical signals, or any suitable combination thereof.

[0092] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present disclosure. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0093] It is to be understood that the above embodiments are merely exemplary embodiments used to illustrate the principles of the present disclosure, but the present disclosure is not limited thereto. For those of ordinary skill in the art, various modifications and improvements can be made without departing from the spirit and substance of the present disclosure, and these modifications and improvements are also considered to be within the scope of protection of the present disclosure.

Claims

1. The automated closed-loop management system for pipeline construction safety hazards based on AI video is characterized by: include: Information collector, used to collect video information of the construction site; An AI server, used to identify illegal behaviors in the video information and generate an alarm signal; The alarm terminal is used to record the violation correction process according to the alarm signal, obtain the violation correction process information, and upload the violation correction process information to the AI ​​server.

2. According to the AI ​​video-based automated closed-loop management system for pipeline construction safety hazards according to claim 1, it is characterized in that: The AI ​​server includes a video recognition module, a violation judgment module and a violation alarm module: The video recognition module is used to identify possible violations in the video information; The violation judgment module is used to judge whether the possible violation behavior is a violation behavior, and send the violation behavior to the violation alarm module; The violation alarm module is used to generate a violation record and an alarm signal according to the violation behavior.

3. The AI ​​video-based automated closed-loop management system for pipeline construction safety hazards according to claim 2 is characterized in that: The video recognition module comprises: A multi-type single model recognition layer, used to identify possible violation types in the video information; The single-type single-model recognition layer is used to call the corresponding recognition model according to the violation type to identify possible violation behaviors.

4. The AI ​​video-based automated closed-loop management system for pipeline construction safety hazards according to claim 2 is characterized in that: The violation judgment module is used to judge whether the possible violation is a violation according to the number of times the possible violation occurs within a preset time range.

5. According to claim 2, the AI ​​video-based automated closed-loop management system for pipeline construction safety hazards is characterized in that: The violation judgment module is used to judge whether multiple violation behaviors are of the same type within the rectification time range. If they are of the same type, the violation judgment module only sends the first violation behavior of the same type to the violation alarm module.

6. The automated closed-loop management system for pipeline construction safety hazards based on AI video according to any one of claims 1 to 5, characterized in that: Also includes: A communication module is used to realize communication between the AI ​​server and the alarm terminal.

7. The automated closed-loop management method for pipeline construction safety hazards based on AI video is characterized by: include: Collect video information from the construction site; Identifying illegal behaviors in the video information and generating an alarm signal; The violation correction process is recorded according to the alarm signal to obtain violation correction process information, and the violation correction process information is uploaded.

8. The method for automated closed-loop management of pipeline construction safety hazards based on AI video according to claim 7 is characterized in that: The step of identifying the illegal behavior in the video information and generating an alarm signal comprises: Identifying possible traffic violations in the video information; Determining whether the possible violation is a violation; A violation record and an alarm signal are generated according to the violation behavior.

9. An electronic device, characterized in that: include: one or more processors; A storage unit, used to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors can implement the automated closed-loop management method for pipeline construction safety hazards based on AI video according to any one of claims 7 to 8.

10. A computer storage medium, characterized in that: Computer-readable instructions are stored, and when the computer-readable instructions are executed by a processor, the automated closed-loop management method for safety hazards in pipeline construction based on AI video as described in any one of claims 7 to 8 is implemented.

Images