Factory unsafe behavior identification method and interlocking alarm system

Abstract

The invention relates to a factory unsafe behavior identification method and an interlocking alarm system, and the method comprises the steps: obtaining a network camera transmission data package, and carrying out the analysis to obtain a factory real-time image; generating an identification result based on the real-time picture; and judging whether the identification result meets at least one preset alarm condition, and if yes, calling a system interlocking strategy corresponding to the preset alarm condition. The system comprises a network camera, a safety behavior recognition platform and a controller, the safety behavior recognition platform is provided with an image processor, a media processing platform and a safety behavior model, and the safety behavior model is provided with a safety behavior recognition algorithm. According to the invention, an intelligent video behavior analysis function is realized, an enterprise is helped to quickly find out illegal operation conditions in a production operation process, safe operation is stopped and ensured in time, and normalization of safety management work is promoted. Passive tracing of factory safety production management is changed into active prevention, an accurate, intelligent and integrated safety management system is realized, and cost reduction and benefit improvement of enterprises are finally realized.

Description

Technical Field

[0001] The present application relates to the technical field of safe production management, and in particular to a method for identifying unsafe behaviors in factories and an interlocking alarm system. Background Art

[0002] In manufacturing plants, continuous production leads to a high probability of unsafe human behavior. Currently, the control of unsafe human behavior can only rely on ideological education and spot checks by safety managers. However, in the actual production process, workers often take risks and violate regulations in order to catch up with the production schedule. The spot checks by safety managers have a probability problem, and it is difficult to control and stop violations throughout the process, which brings great safety hazards.

[0003] It is necessary to invent a method for identifying unsafe behaviors in factories, identify unsafe behaviors, and realize timely shutdown of dangerous equipment through an associated alarm system. Summary of the invention

[0004] In order to solve or partially solve the problems existing in the related technologies, the present application provides a factory unsafe behavior identification method and interlocking alarm system, which adopts the "video surveillance + AI visual recognition + algorithm" approach to realize the intelligent capture, identification and alarm of unsafe behaviors of people in dangerous positions throughout the entire process, and automatically interlocks and stops machinery and equipment that may cause harm to people based on the system.

[0005] The first aspect of the present application provides a method for identifying unsafe behaviors in a factory, comprising the following steps:

[0006] Acquire the data packets transmitted by the network camera and parse them to get the real-time images of the factory. There are several network cameras installed in the factory and the image acquisition area covers the risk area of ​​the factory;

[0007] Generate recognition results based on real-time images, including at least one of whether a person enters the area under the zinc stripping knife, whether a person crosses the mobile trolley in the hanging board area, whether a person enters the zinc sheet stacking area, and whether a person enters the area under the overhead crane;

[0008] It is determined whether the identification result satisfies at least one preset alarm condition. If so, the system interlocking strategy corresponding to the preset alarm condition is called. The system interlocking strategy includes the alarm of the on-site alarm and / or the shutdown of the zinc stripping machine.

[0009] In an implementation of the first aspect, the preset alarm condition includes:

[0010] When the first recognition result is that a person enters the area under the zinc stripping knife, a zinc stripping machine shutdown signal is triggered.

[0011] In an implementation of the first aspect, the preset alarm condition includes:

[0012] When the first recognition result is that there is a person crossing the mobile trolley in the hanging board area, the zinc stripping machine stop signal is triggered.

[0013] In an implementation of the first aspect, the preset alarm condition includes:

[0014] When the first recognition result is that a person enters the zinc sheet stacking area, a zinc stripping machine shutdown signal is triggered.

[0015] In an implementation of the first aspect, the preset alarm condition includes:

[0016] When the second recognition result is that a person has entered the area under the overhead crane, an alarm signal of the on-site alarm is triggered.

[0017] The first aspect of the present application provides a factory unsafe behavior interlock alarm system, which is applicable to the factory unsafe behavior identification method shown in the first aspect, and includes:

[0018] A network camera, connected to an image processor, is used to collect real-time images of the factory;

[0019] A safety behavior recognition platform is provided with an image processor, a media processing platform and a safety behavior model. The safety behavior model is provided with a safety behavior recognition algorithm. The safety behavior recognition platform is used to monitor and analyze personnel behavior in real time and generate control signals;

[0020] The controller is connected with the alarm and the zinc stripping machine, and switches the alarm and the zinc stripping machine on and off based on the control signal.

[0021] In one implementation of the second aspect, the interlocking alarm system is matched with the factory risk area. Multiple interlocking alarm systems are installed in the factory. The interlocking alarm system is connected to the cloud control platform. The cloud control platform is used to read the system data of each interlocking alarm system to realize intelligent scheduling and data decision analysis of each risk area.

[0022] The technical solution provided by this application may have the following beneficial effects:

[0023] This application provides a method for identifying unsafe behaviors in factories and an interlocking alarm system, which realizes the intelligent video behavior analysis function, helps enterprises quickly discover illegal operations in the production process, so as to stop them in time to ensure safe operations and promote the normalization of safety management. It transforms factory safety production management from traditional passive tracing to active prevention, realizes a precise, intelligent, and integrated safety management system, and ultimately achieves cost reduction and efficiency improvement for enterprises.

[0024] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The above and other objects, features and advantages of the present application will become more apparent through a more detailed description of exemplary embodiments of the present application in conjunction with the accompanying drawings, wherein the same reference numerals generally represent the same components in the exemplary embodiments of the present application.

[0026] Figure 1 It is a flow chart of a method for identifying unsafe behaviors in a factory shown in an embodiment of the present application;

[0027] Figure 2 is a schematic diagram of the connection between the interlock alarm system and the cloud control platform shown in the embodiment of the present application;

[0028] Figure 3 It is a schematic diagram of the operation of the "edge cloud" algorithm shown in the embodiment of the present application. DETAILED DESCRIPTION

[0029] The embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although the embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to make the present application more thorough and complete, and to fully convey the scope of the present application to those skilled in the art.

[0030] It should be understood that although the terms "first", "second", "third", etc. may be used in this application to describe various information, this information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of this application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of this application, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0031] In the description of the present application, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0032] Unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0033] The technical solution of the embodiments of the present application is described in detail below with reference to the accompanying drawings.

[0034] Embodiment 1

[0035] A factory unsafe behavior interlock alarm system includes: a network camera, a safe behavior recognition platform and a controller. The network camera is connected to an image processor to collect real-time factory images. Domestic mainstream camera suppliers are preferred. According to the needs of scene applications, 1080P, 2K, 4K and even 8K ultra-high-definition video monitoring requirements can be met. At the same time, thermal imaging cameras are selected for special scene needs.

[0036] The security behavior recognition platform is equipped with an image processor, a media processing platform and a security behavior model. The security behavior model is equipped with a security behavior recognition algorithm. The security behavior recognition platform is used to monitor and analyze personnel behavior in real time and generate control signals. The image processor uses an edge computing server, and takes advantage of the powerful processing and analysis capabilities and compatibility of the edge to perform calculations and analysis on the client side or data edge, and transmit the results back to the background, greatly reducing transmission costs and central computer room construction costs.

[0037] The controller is connected with the alarm and the zinc stripping machine, and switches the alarm and the zinc stripping machine on and off based on the control signal.

[0038] like Figure 2 As shown in the figure, the interlock alarm system is matched with the factory risk area. There are multiple interlock alarm systems in the factory. The interlock alarm system is connected to the cloud control platform. The cloud control platform is used to read the system data of each interlock alarm system to realize the intelligent scheduling and data decision analysis of each risk area. Each risk area uses independent edge intelligent computing and storage. The headquarters cloud control platform only reads the "N" system data on site and does not participate in the on-site system algorithm. The headquarters cloud control platform realizes intelligent scheduling and data decision analysis of N on-site scenes, which is not limited by the network performance between the group and the site, and reduces the computing capacity load of the cloud computing center of the group headquarters.

[0039] The safety behavior recognition platform uses cameras, sensors and other equipment to monitor and analyze people's behavior and safety-related events in real time through data analysis, artificial intelligence, and machine learning. The safety behavior recognition platform has a mature application platform and related tools that can adapt to various types of cameras and sensors, while meeting the API interface that can be quickly delivered to the upper level, and can achieve the enterprise's hierarchical management requirements and realize the intelligent closed-loop management of enterprise safety production. The deployment and application of safety behavior recognition algorithms should be fast and convenient, shortening the delivery time and improving the efficiency of algorithm deployment. This requires the algorithm supplier to also have a mature comprehensive management and control platform that can train, deploy and deliver integrated services.

[0040] After the algorithm is developed, on-site visual recognition training is carried out. The training is to let employees practice illegal behaviors in the camera area on site, let the system capture no less than 30 times / types of illegal behaviors and compare them with the standard algorithm in the system. After multiple comparisons, the system records and analyzes the illegal behaviors, thereby continuously improving the accuracy of illegal behavior capture and analysis to more than 95%.

[0041] Standardize the definition of unsafe behaviors of people in dangerous positions, and through AI visual recognition, analysis, and training, the visual recognition accuracy rate reaches more than 95%; through the "edge cloud" architecture, a "1+N" video violation capture system solution of "edge smart server + industrial customized algorithm model + comprehensive intelligent management and control platform" is proposed. "1" is a cloud control platform, and "N" is N on-site intelligent protection interlocking alarm systems. Independent edge intelligent computing and storage of N scenes on site are realized, and the headquarters cloud control platform realizes intelligent scheduling and data decision analysis of N on-site scenes, which is not limited by the network performance between the group and the site, and reduces the computing capacity load of the cloud computing center of the group headquarters.

[0042] like Figure 3 As shown in the figure, during the entire algorithm application process of the "edge cloud", the streaming media server of the system platform obtains the data packet transmitted by the network camera by parsing the RTSP protocol, and then the media processing platform parses the data packet into a real-time image (picture), and inputs the picture into the model algorithm for pre-processing and detection and recognition calculation of the picture; then the detection and recognition result is added to the alarm logic judgment, where the alarm logic judgment includes: boundary judgment, location recognition, alarm object judgment, alarm time judgment, event duration, etc.; then the alarm event information is pushed to the intelligent analysis system, which manages the analysis box, camera and other related hardware and algorithms, alarm information, etc.; finally, the intelligent analysis system pushes the alarm event to the management and control cloud platform or links the peripherals for event alarm.

[0043] Embodiment 2

[0044] like Figure 1The method for identifying unsafe behaviors in a factory shown includes the following steps:

[0045] S1. Obtain the data packet transmitted by the network camera and parse it to obtain the real-time picture of the factory. There are several network cameras set up in the factory and the image acquisition area covers the risk area of ​​the factory.

[0046] After implementing the risk identification of zinc stripping machine positions, four major safety risk points were summarized: ① People cannot pass under the zinc stripping knife; ② People cannot cross the mobile trolley in the hanging board area; ③ People cannot enter the zinc sheet stacking area during the operation of the machine; ④ People cannot stand under the overhead crane. Corresponding on-site cameras were installed to ensure that the unsafe behaviors of people at the above identified risk points can be monitored 360 degrees.

[0047] S2. Generate recognition results based on the real-time image, including at least one of whether a person enters the area under the zinc stripping knife, whether a person crosses the mobile trolley in the hanging area, whether a person enters the zinc sheet stacking area, and whether a person enters the area under the overhead crane.

[0048] After implementing the four risk identifications at the zinc stripping machine position, the judgment criteria were determined respectively: ① When a person passes under the zinc stripping knife, a signal is directly given to stop the zinc stripping machine; ② When a person crosses over the mobile trolley in the hanging board area, a signal is directly given to stop the zinc stripping machine; ③ When a person enters the zinc sheet stacking area, a signal is directly given to stop the zinc stripping machine; ④ When a person stands under the overhead crane, the on-site alarm sounds to warn personnel to leave the danger zone as soon as possible.

[0049] S3. Determine whether the identification result satisfies at least one preset alarm condition. If so, call the system interlocking strategy corresponding to the preset alarm condition. The system interlocking strategy includes on-site alarm alarm and / or zinc stripping machine shutdown.

[0050] After determining the unsafe behavior, analyze the system interlocking strategy that caused the behavior. There are two main interlocking strategies: one is the on-site alarm alarm, which warns employees to stop the unsafe behavior in time; the other is the system sends a signal to the DCS or PLC to directly shut down the running equipment.

[0051] The preset alarm conditions include:

[0052] When the first recognition result is that a person enters the area under the zinc stripping knife, a zinc stripping machine stop signal is triggered;

[0053] When the first recognition result is that there is a person crossing the mobile trolley in the hanging board area, the zinc stripping machine stop signal is triggered;

[0054] When the first recognition result is that a person enters the zinc sheet stacking area, a zinc stripping machine stop signal is triggered;

[0055] When the second recognition result is that a person has entered the area under the overhead crane, an alarm signal of the on-site alarm is triggered.

[0056] Through AI visual recognition, the system automatically determines unsafe human behavior; after AI intelligently determines unsafe human behavior, it transmits the signal to the factory DCS control platform. The control platform issues instructions. First, it activates the alarm system to remind the operator that unsafe behavior has occurred and the behavior should be stopped in time; second, it synchronously interlocks the operator to shut down nearby equipment to achieve the purpose of inherent safety.

[0057] Finally, it should be noted that, in this article, relationships such as first and second, etc. are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms include, include or any other variations are intended to cover non-exclusive inclusion, so that a process, method, article or device that includes a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0058] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0059] The embodiments of the present application have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A method for identifying unsafe behaviors in a factory, characterized in that: The following steps are involved: Acquire the data packets transmitted by the network camera and parse them to obtain the real-time images of the factory. The network camera is provided in a plurality of locations in the factory and the image acquisition area covers the risk area of ​​the factory. Based on the real-time picture, a recognition result is generated, and the recognition result includes at least one of whether a person enters the area under the zinc stripping knife, whether a person crosses the mobile trolley in the hanging board area, whether a person enters the zinc sheet stacking area, and whether a person enters the area under the overhead crane; It is determined whether the identification result satisfies at least one preset alarm condition. If so, a system interlocking strategy corresponding to the preset alarm condition is called, wherein the system interlocking strategy includes an on-site alarm and / or a zinc stripping machine shutdown.

2. The method for identifying unsafe behaviors in factories according to claim 1, characterized in that: The preset alarm conditions include: When the first recognition result is that a person enters the area under the zinc stripping knife, a zinc stripping machine shutdown signal is triggered.

3. The method for identifying unsafe behaviors in factories according to claim 1, characterized in that: The preset alarm conditions include: When the first recognition result is that there is a person crossing the mobile trolley in the hanging board area, a stop signal of the zinc stripping machine is triggered.

4. The method for identifying unsafe behaviors in factories according to claim 1, characterized in that: The preset alarm conditions include: When the first recognition result is that a person enters the zinc sheet stacking area, a zinc stripping machine shutdown signal is triggered.

5. The method for identifying unsafe behaviors in factories according to claim 1, characterized in that: The preset alarm conditions include: When the second recognition result is that a person has entered the area under the overhead crane, an alarm signal of the on-site alarm is triggered.

6. A factory unsafe behavior interlock alarm system, applicable to the factory unsafe behavior identification method according to any one of claims 1 to 5, characterized in that: include: A network camera, connected to an image processor, is used to collect real-time images of the factory; A safety behavior recognition platform is provided with an image processor, a media processing platform and a safety behavior model, wherein the safety behavior model is provided with a safety behavior recognition algorithm, and the safety behavior recognition platform is used to monitor and analyze the behavior of personnel in real time and generate control signals; The controller is connected with the alarm and the zinc stripping machine, and switches the alarm and the zinc stripping machine based on the control signal.

7. The factory unsafe behavior interlock alarm system according to claim 6, characterized in that: The interlock alarm system is matched with the factory risk area. A plurality of the interlock alarm systems are arranged in the factory. The interlock alarm system is connected to a cloud control platform. The cloud control platform is used to read the system data of each interlock alarm system to realize intelligent scheduling and data decision analysis of each risk area.

Images