Power plant unsafe behavior platform-based violation operation detection method and device for fire operation scene, storage medium and processor

Abstract

The invention discloses a violation operation detection method and device for a fire operation scene based on a power plant unsafe behavior platform, a storage medium and a processor, relates to the technical field of operation safety management, and mainly aims to accurately detect and early warn violation operation and reduce potential safety hazards in the fire operation scene. According to the main technical scheme, the method comprises the steps of obtaining at least one target image frame corresponding to a live working area, wherein the target image frame is used for representing video image frames continuously intercepted in a video stream of the live working area at a preset frequency; inputting the at least one target image frame into a pre-trained target detection model to obtain category information and coordinate information of each target in the at least one target image frame; according to the category information and the coordinate information, whether violation operation exists in the fire operation area or not is judged; and if yes, executing a corresponding alarm operation based on the hidden danger level of the violation operation.

Description

Technical Field

[0001] The present application relates to the technical field of operation safety management, and in particular to a method, device, storage medium and processor for detecting illegal operations in hot work scenarios based on an unsafe behavior platform of a power plant. Background Art

[0002] In the power industry, especially in places such as gas-fired thermal power plants, hot work is a high-risk activity involving welding, cutting or other spark-generating work. Since such work may cause fire or explosion, posing a serious threat to personnel and facilities, strict safety regulations must be followed. The safety management of hot work mainly includes two stages: preparation before the operation and monitoring during the operation. Real-time monitoring of these two stages is one of the key measures to ensure the safety of hot work.

[0003] At present, the detection of illegal operations in hot work scenarios mainly relies on traditional video surveillance systems to collect image frames during hot work, and use models to analyze whether the equipment or personnel contained in the image frames are complete to control the alarm. However, hot work scenarios may occur at any time in any location in places such as thermal power plants. In addition to the need for complete equipment or personnel, some specific illegal operations need to be intelligently analyzed, especially in terms of equipment configuration layout, work supervisor absence, etc. The above method cannot accurately detect and warn these specific illegal operations, resulting in an increase in safety hazards in hot work scenarios. Summary of the invention

[0004] In view of the above problems, the present application provides a method, device, storage medium and processor for detecting illegal operations in hot work scenarios based on the power plant unsafe behavior platform. The main purpose is to accurately detect and warn of illegal operations and reduce safety hazards in hot work scenarios.

[0005] In order to solve the above technical problems, this application proposes the following solutions:

[0006] In a first aspect, the present application provides a method for detecting illegal operations in a hot work scenario based on an unsafe behavior platform of a power plant, the method comprising:

[0007] Acquire at least one target image frame corresponding to a hot work area, wherein the hot work area is a monitoring area where a mobile control ball is fixed at a designated position to shoot the hot work, and the target image frame is used to represent video image frames continuously captured at a preset frequency in a video stream of the hot work area;

[0008] Inputting at least one of the target image frames into a pre-trained target detection model to obtain category information and coordinate information of each target in at least one of the target image frames;

[0009] Determine whether there is any illegal operation in the hot work area according to the category information and the coordinate information;

[0010] If so, a corresponding warning operation is performed based on the risk level of the illegal operation.

[0011] In a second aspect, the present application provides a device for detecting illegal operations in a hot work scenario based on an unsafe behavior platform of a power plant, the device comprising:

[0012] An acquisition unit is used to acquire at least one target image frame corresponding to a hot work area, wherein the hot work area is a monitoring area where a mobile control ball is fixed at a designated position to shoot the hot work, and the target image frame is used to represent video image frames continuously captured at a preset frequency in a video stream of the hot work area;

[0013] a processing unit, configured to input the at least one target image frame obtained by the acquisition unit into a pre-trained target detection model to obtain category information and coordinate information of each target in the at least one target image frame;

[0014] A judging unit, configured to judge whether there is any illegal operation in the hot work area according to the category information and the coordinate information obtained by the processing unit;

[0015] The alarm unit is used to perform corresponding alarm operations based on the hidden danger level of the illegal operation if the judgment unit judges that there is illegal operation in the hot work area.

[0016] In order to achieve the above-mentioned purpose, according to the third aspect of the present application, a storage medium is provided, wherein the storage medium includes a stored program, wherein when the program is running, the device where the storage medium is located is controlled to execute the above-mentioned first aspect of the illegal operation detection method for the hot work scenario based on the power plant unsafe behavior platform.

[0017] In order to achieve the above-mentioned purpose, according to the fourth aspect of the present application, a processor is provided, which is used to run a program, wherein when the program is running, the illegal operation detection method for the hot work scenario based on the power plant unsafe behavior platform of the above-mentioned first aspect is executed.

[0018] Through the above-mentioned technical scheme, the present application provides a method, device, storage medium and processor for detecting illegal operations in hot work scenarios based on the power plant unsafe behavior platform. When it is necessary to detect illegal operations in hot work scenarios based on the power plant unsafe behavior platform, at least one target image frame continuously captured in the video stream of the hot work area at a preset frequency is obtained, and at least one target image frame is input into a pre-trained target detection model to obtain category information and coordinate information of each target in the at least one target image frame, and it is judged whether there is any illegal operation in the hot work area according to the category information and coordinate information. If so, a corresponding alarm operation is performed based on the hidden danger level of the illegal operation. Through the technical solution provided in this application, it is possible to continuously identify the category and coordinates of each target in the hot work area, and to comprehensively detect whether there are any specific illegal operations in the hot work area involving the configuration layout of the equipment, the absence of the work supervisor, etc., through the category and coordinates of each target, and to execute an alarm according to the hidden danger level of the illegal operation, especially when the hot work scene changes its location and needs to be redeployed, the accuracy of detection and early warning of illegal operations is effectively improved, thereby reducing the safety hazards in the hot work scene, and providing more reliable safety protection for personnel and facilities in the hot work scene.

[0019] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present application. Also, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0021] Figure 1 A flow chart of a method for detecting illegal operations in a hot work scenario based on an unsafe behavior platform of a power plant provided in an embodiment of the present application is shown;

[0022] Figure 2 A flow chart of another method for detecting illegal operations in a hot work scenario based on an unsafe behavior platform of a power plant provided in an embodiment of the present application is shown;

[0023] Figure 3 A block diagram showing a composition of a device for detecting illegal operations in a hot work scenario based on an unsafe behavior platform of a power plant provided in an embodiment of the present application is shown;

[0024] Figure 4 A block diagram of another illegal operation detection device for hot work scenarios based on an unsafe behavior platform of a power plant provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0025] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0026] At present, the detection of illegal operations in hot work scenarios mainly relies on traditional video surveillance systems to collect image frames during hot work, and use models to analyze whether the equipment or personnel contained in the image frames are complete to control the alarm. However, hot work scenarios may occur at any time in any location in places such as thermal power plants. In addition to the need for complete equipment or personnel, some specific illegal operations need to be intelligently analyzed, especially in terms of equipment configuration layout, work supervisor absence, etc. The above method cannot accurately detect and warn these specific illegal operations, resulting in an increase in safety hazards in hot work scenarios.

[0027] After research, the inventor found that it is possible to continuously capture corresponding image frames from the video stream of the action operation area, and use the model to identify the category and coordinates of each target in the image frame. The category and coordinates are used to comprehensively judge whether there are specific illegal operations in the action operation area, such as equipment configuration layout, and the absence of the work supervisor, and to issue an alarm based on the hidden danger level of the illegal operation. In this way, the category and coordinates are used to reflect the presence and location layout of equipment or personnel in the action operation area, which effectively improves the accuracy of detection and early warning of illegal operations, thereby reducing the safety hazards in the hot work scene and providing more reliable safety protection for personnel and facilities in the hot work scene.

[0028] Based on the above considerations, the embodiment of the present application provides a method for detecting illegal operations in hot work scenarios based on the unsafe behavior platform of a power plant. This method can accurately detect and warn illegal operations, thereby reducing safety hazards in hot work scenarios. The specific execution steps are as follows: Figure 1 As shown, including:

[0029] 101. Acquire at least one target image frame corresponding to a hot work area.

[0030] The hot work area is a monitoring area where a mobile control ball is fixed at a designated position to shoot hot work, and the target image frame is used to represent the video image frames continuously captured in the video stream of the hot work area at a preset frequency.

[0031] This embodiment is applied to the power plant unsafe behavior platform, which can be pre-connected with the mobile control ball. After determining the operation point of the action operation, the mobile control ball is installed at the designated location of the hot work area. At this designated location, it can ensure that the monitoring area of ​​the mobile control ball shooting the hot work covers the entire operation point. The mobile control ball should have high definition and wide-angle vision to capture the full picture of the hot work area.

[0032] In this step, after the mobile control ball is turned on, it will record the real-time video stream of the hot work area and transmit the real-time video stream to the power plant unsafe behavior platform. The power plant unsafe behavior platform continuously captures the corresponding video image frames in the real-time video stream at a preset frequency to obtain at least one target video stream. Among them, the preset frequency is specifically the interception frequency, which can be customized according to the monitoring accuracy of the hot work, such as 10 frames per second, 5 frames per second, etc. It should be noted that the interception frequency must also match the acquisition frequency of the real-time video stream to ensure that the target image frame accurately reflects the real-time status of the action operation area.

[0033] 102. Input at least one target image frame into a pre-trained target detection model to obtain category information and coordinate information of each target in the at least one target image frame.

[0034] The coordinate information is the coordinates of the frame of the object selected by the bounding box.

[0035] Advanced target detection models are used in advance to train key targets in hot work. The target detection model can be one of the models such as YOLOv8, Faster R-CNN or SSD. The key targets include but are not limited to work supervisors, handheld fire extinguishers, oxygen cylinders, acetylene cylinders, etc. Among them, oxygen cylinders and acetylene cylinders are commonly used gas sources in hot work. Handheld fire extinguishers are used to quickly extinguish fires in the early stages, while work supervisors are responsible for supervising the hot work process, which is used to stop unsafe behaviors in time and take necessary emergency measures in emergency situations. Specifically, a large amount of image data containing hot work scenes under various conditions is first collected. The various conditions can be different lighting, weather, time or environment. It is particularly noted that the image data needs to contain key targets such as handheld fire extinguishers, oxygen cylinders, acetylene cylinders and work supervisors. The collected image data is annotated according to the key targets, that is, the positions of handheld fire extinguishers, oxygen cylinders, acetylene cylinders and work supervisors are marked, and bounding boxes are usually used to frame them to ensure that the target detection model can accurately learn the features of these key targets. Select a suitable deep learning model from models such as YOLOv8, Faster R-CNN or SSD as the target detection model, and use the labeled image data to train the selected target detection model so that it can simultaneously locate and classify multiple targets in the image data in a single detection, that is, output the category information and coordinate information of each target. The coordinate information is the coordinates of the bounding box of the target selected.

[0036] In this step, at least one target image frame is input into a trained target detection model, which will identify the various targets contained in the at least one target image frame and output the category label and bounding box coordinates for each target, that is, obtain the category information and coordinate information of each target in the at least one target image frame.

[0037] 103. Determine whether there are any illegal operations in the hot work area based on the category information and coordinate information.

[0038] In practical applications, since the safety management of action operations mainly targets the two stages of preparation before the operation and monitoring during the operation, in this step, a set of compliance operation rules corresponding to the hot work scene can be pre-defined, including but not limited to the simultaneous presence of work guardians, handheld fire extinguishers, oxygen cylinders and acetylene cylinders in the hot work area, whether the oxygen cylinder has a safety helmet and a shockproof ring, whether the distance between the oxygen cylinder and the acetylene cylinder is greater than the preset distance, and the duration of the work guardian's absence from the post is less than the preset duration. This embodiment does not limit this. Among them, for the simultaneous presence of work guardians, handheld fire extinguishers, oxygen cylinders and acetylene cylinders in the hot work area, it can be determined by the corresponding category information in any target image frame, and the work guardian needs to walk around and inspect in the hot work scene, and it is allowed to leave the post in a short time. Therefore, it is necessary to determine the corresponding category information in multiple continuous target image frames, and the handheld fire extinguisher, oxygen cylinder and acetylene cylinder are now deployed based on the determination of the action operation area, and they may be replaced or moved in the middle. Therefore, it is necessary to determine the category information and coordinate information corresponding to at least one target image frame together in real time.

[0039] By judging the above-mentioned compliance operation rules through the category information and coordinate information corresponding to at least one target image frame or multiple consecutive image frames, it can be determined whether there are any illegal operations in the hot work area. If not, it means that the action operation area currently meets the compliance operation rules. If so, it means that the action operation area currently does not meet the compliance operation rules. At this time, step 104 can be executed.

[0040] 104. Perform corresponding warning operations based on the risk level of the illegal operation.

[0041] In this step, the specific illegal operation can be determined from step 103, but since the severity of the safety hazards caused by different illegal operations is different, different illegal operations can be divided into different hazard levels according to the severity, such as preparation hazard level and safety hazard level, among which the preparation hazard level corresponds to the presence of a work supervisor, a handheld fire extinguisher, an oxygen cylinder and an acetylene cylinder in the hot work area in the aforementioned step 103, and the safety hazard level corresponds to whether the oxygen cylinder in the aforementioned step 103 is equipped with a safety helmet and a shockproof ring, whether the distance between the oxygen cylinder and the acetylene cylinder is greater than the preset distance, and whether the duration of the work supervisor's absence from his post is less than the preset duration, etc.

[0042] It should be noted that, since the preparation hazard level usually occurs before the operation, while the safety hazard level usually occurs during the operation, the preparation hazard level has a lower degree of urgency in its alarm operation than the safety hazard level. In order to enable relevant personnel to effectively distinguish alarms for different hazard levels, corresponding alarm strategies can be formulated for each hazard level. For example, the preparation hazard level can trigger SMS or email notifications to the relevant person in charge of the hot work, while the safety hazard level can immediately trigger an audible and visual alarm and notify all relevant operators to stop the operation, etc., to ensure timely response and reduce safety hazards.

[0043] Based on the above Figure 1 It can be seen from the implementation method that a method for detecting illegal operations in hot work scenarios based on the power plant unsafe behavior platform is provided. When it is necessary to detect illegal operations in hot work scenarios based on the power plant unsafe behavior platform, at least one target image frame continuously captured in the video stream in the hot work area at a preset frequency is obtained, and at least one target image frame is input into a pre-trained target detection model to obtain category information and coordinate information of each target in at least one target image frame, and it is judged whether there is any illegal operation in the hot work area according to the category information and coordinate information. If so, a corresponding alarm operation is performed based on the hidden danger level of the illegal operation. Through the technical solution provided in this application, it is possible to continuously identify the category and coordinates of each target in the hot work area, and to comprehensively detect whether there are any specific illegal operations in the hot work area involving the configuration layout of the equipment, the absence of the work supervisor, etc., through the category and coordinates of each target, and to execute an alarm according to the hidden danger level of the illegal operation, especially when the hot work scene changes its location and needs to be redeployed, the accuracy of detection and early warning of illegal operations is effectively improved, thereby reducing the safety hazards in the hot work scene, and providing more reliable safety protection for personnel and facilities in the hot work scene.

[0044] Further, the preferred embodiment of the present application is in the above Figure 1 Based on this, a detailed description of the process of illegal operation detection in hot work scenarios based on the power plant unsafe behavior platform is given. The specific steps are as follows: Figure 2 As shown, including:

[0045] 201. Acquire at least one target image frame corresponding to a hot work area.

[0046] This step is combined with the description of step 101 in the above method, and the same contents will not be repeated here.

[0047] 202. Input at least one target image frame into a pre-trained target detection model to obtain category information and coordinate information of each target in the at least one target image frame.

[0048] This step is combined with the description of step 102 in the above method, and the same contents are not repeated here.

[0049] 203. Obtain the preset compliance operation rules for the hot work area.

[0050] Among them, the compliance operation rules include pre-rules and post-rules. The pre-rules are used to determine whether the scene preparation in the hot work area is compliant, and the post-rules are used to determine whether the scene safety in the hot work area is compliant.

[0051] In this step, the compliance operation rules can be formulated by a professional safety team in combination with industry standards and the particularity of hot work. The compliance operation rules at least include but are not limited to the simultaneous presence of work supervisors, handheld fire extinguishers, oxygen cylinders and acetylene cylinders in the hot work area, whether the oxygen cylinders are equipped with safety helmets and shockproof rings, whether the distance between oxygen cylinders and acetylene cylinders is greater than the preset distance, and the duration of the work supervisor's absence from work is less than the preset duration, etc. The compliance operation rules can be pre-stored in the rule base of the power plant unsafe behavior platform, which can be divided according to different operation scenarios, and can be obtained from the rule base when performing illegal operation detection in the hot work scenario later.

[0052] It should be noted that since the safety management of hot work usually targets two stages: preparation before the work and monitoring during the work, in order to avoid directly testing the safety during the work when the preparation before the work is already insufficient, the compliance work rules can be divided into pre-rules and post-rules. The pre-rules are used to determine whether the scene preparation in the hot work area is compliant, that is, to focus on the preparation before the work, and the post-rules are used to determine whether the scene safety in the hot work area is compliant, that is, to focus on the safety during the work. Among them, the pre-rules specifically refer to the simultaneous presence of work supervisors, handheld fire extinguishers, oxygen cylinders and acetylene cylinders in the hot work area, and the post-rules at least include whether the oxygen cylinder has a safety helmet and a shockproof ring, whether the distance between the oxygen cylinder and the acetylene cylinder is greater than the preset distance, and whether the work supervisor's absence lasts less than the preset time. By dividing the compliance work rules according to the work process, not only can the detection process be more logical and targeted, but it can also avoid directly testing the safety during the work when the preparation before the work is already insufficient, thereby reducing the waste of computing resources.

[0053] 204. Determine whether the hot work area meets the pre-conditions based on the category information.

[0054] Among them, the pre-rules include multiple necessary categories corresponding to the scene preparation.

[0055] From the above steps, it can be seen that the pre-rule specifically refers to the simultaneous presence of work supervisors, handheld fire extinguishers, oxygen cylinders, and acetylene cylinders in the hot work area. In this step, in order to make subsequent alarms more targeted based on the judgment results of the pre-rules, work supervisors, handheld fire extinguishers, oxygen cylinders, and acetylene cylinders can be used as multiple necessary categories corresponding to the scene preparation. By comparing it with the category information, it can be determined whether the hot work area meets the pre-rule.

[0056] The specific implementation process of judging whether the hot work area meets the pre-rules based on the category information is as follows: judging whether there are work supervisors, hand-held fire extinguishers, oxygen cylinders and acetylene cylinders in the hot work area at the same time based on the category information; if so, it is determined that the hot work area meets the pre-rules; if not, it is determined that the hot work area does not meet the pre-rules.

[0057] Since the category information is obtained by analyzing each target contained in the target image frame through the target detection model, by comparing the category of each target with the necessary category, it can be determined whether all necessary categories exist. If so, it is considered that the pre-rule is satisfied, and step 206 is executed to further judge the post-rule. Otherwise, it is considered that the pre-rule is not satisfied, and step 205 is executed.

[0058] 205. Determine the illegal operations with the first hidden danger level in the hot work area, and screen out the non-compliant categories from multiple necessary categories.

[0059] Among them, the first hidden danger level is used to characterize the preparation hidden danger.

[0060] In this step, in order to make the subsequent alarms more targeted at the illegal operations corresponding to the preparation hidden dangers, the illegal operations can be divided into different hidden danger levels, and the first hidden danger level corresponds to the preparation hidden dangers. When it is detected that the pre-rules are not satisfied, there are illegal operations of the first hidden danger level. In addition, according to the comparison results of the categories of each target and the necessary categories in the aforementioned step 204, the substandard categories in the necessary categories, that is, the missing categories in the necessary categories, can be obtained, so that the subsequent alarms can generate corresponding violation information based on the substandard categories. By defining the hidden danger levels and generating specific violation information, the subsequent alarms can be made more specific and accurate, that is, more targeted, so that the relevant person in charge can quickly identify the problem and take remedial measures in time.

[0061] 206. Determine whether the hot work area meets the post-rules based on the category information and coordinate information.

[0062] Among them, the post-rules include multiple necessary conditions corresponding to scene safety.

[0063] It can be seen from the above steps that the post-rules at least include whether the oxygen cylinder has a safety helmet and a shockproof ring, whether the distance between the oxygen cylinder and the acetylene cylinder is greater than the preset distance, and the duration of the work supervisor's absence from the post is less than the preset time. In this step, in order to facilitate the subsequent alarms to be more targeted according to the judgment results of the post-rules, whether the oxygen cylinder has a safety helmet and a shockproof ring, whether the distance between the oxygen cylinder and the acetylene cylinder is greater than the preset distance, and the duration of the work supervisor's absence from the post is less than the preset time can be used as multiple necessary conditions corresponding to scene safety. By judging each necessary condition separately through category information and coordinate information, it can be determined whether the hot work area meets the post-rules. The specific implementation process of judging whether the hot work area satisfies the post-rules according to the category information and coordinate information is as follows: judging whether the oxygen cylinder is equipped with a safety helmet and a shockproof ring according to the category information, judging whether the duration of the work supervisor's absence from his post is less than the preset duration according to the category information, and judging whether the distance between the oxygen cylinder and the acetylene cylinder is greater than the preset distance according to the coordinate information; if not, it is determined that the hot work area does not satisfy the post-rules; if so, it is determined that the hot work area satisfies the post-rules.

[0064] It should be noted that the safety helmet is mainly used to prevent oxygen leakage, and the shockproof ring is mainly used to reduce physical damage. The two can improve the safety of oxygen cylinders, that is, the safety helmet and the shockproof ring are essential safety accessories for oxygen cylinders. The work supervisor usually wears equipment that can represent his identity, such as work clothes, safety helmets or armbands with specified characteristics, and the specified characteristics include but are not limited to color, style, etc. Therefore, in order to accurately identify the safety helmet, shockproof ring and work supervisor, the safety helmet, shockproof ring and the work clothes, safety helmet or armband corresponding to the work supervisor also exist in the image data used for pre-training the target detection model as one of the key targets.

[0065] Specifically, checking whether the safety helmet and shockproof ring exist in the category information can determine whether the safety of the oxygen cylinder is good. Although the presence of the work supervisor is the key to ensuring the safety of the hot work, he may leave his post for a short time due to walking inspections, that is, he may appear or disappear repeatedly in the target video frame, and this short-term appearance or disappearance is allowed. Based on this, in order to avoid the alarm being too sensitive, it is also necessary to calculate the duration of the continuous absence of the work supervisor in the category information, that is, the duration of the work supervisor's absence from the post. By comparing the duration of absence from the post with the preset duration, it can be determined whether the work supervisor is really off the post. Oxygen cylinders and acetylene cylinders are commonly used gas sources in hot work. Oxygen is used to assist combustion, and acetylene is used to generate high-temperature flames. Both must strictly abide by the safety distance and usage specifications to avoid explosion. Therefore, in addition to the category information, it is also necessary to combine the coordinate information to monitor the location layout of oxygen cylinders and acetylene cylinders, that is, determine the actual distance between the oxygen cylinder and the acetylene cylinder, and compare it with the preset distance to determine whether there is an explosion risk between the oxygen cylinder and the acetylene cylinder.

[0066] Furthermore, since the preset distance refers to a safe distance at the physical level, and the coordinate information refers to the image coordinates in the target video frame, in order to ensure the accuracy of the judgment of the distance between the oxygen cylinder and the acetylene cylinder and the preset distance, before judging whether the distance between the oxygen cylinder and the acetylene cylinder is greater than the preset distance based on the coordinate information, the method also includes: determining the coordinates of the first frame center point corresponding to the oxygen cylinder based on the frame coordinates of the oxygen cylinder; determining the coordinates of the second frame center point corresponding to the acetylene cylinder based on the frame coordinates of the acetylene cylinder; calculating the image distance between the coordinates of the first frame center point and the coordinates of the second frame center point on the horizontal axis of the image; using the distance mapping relationship to convert the image distance into a physical distance, and using the physical distance as the distance between the oxygen cylinder and the acetylene cylinder.

[0067] It should be noted that the frame coordinates specifically refer to the coordinates of the four corner points of the rectangular bounding box. The frame coordinates corresponding to the oxygen cylinder and the acetylene cylinder can be used to determine the coordinates of the center points of the respective frame, that is, the coordinates of the first frame center point and the second frame center point. Since the target video frame is a two-dimensional image, the size of the oxygen cylinder and the acetylene cylinder in the target video frame may be different, that is, the coordinates of the first frame center point and the coordinates of the second frame center point may not be at the same height, that is, not on the same image horizontal axis. Therefore, in order to accurately determine the horizontal distance between the coordinates of the first frame center point and the coordinates of the second frame center point, the number of pixels of the first frame center point and the second frame center point in the horizontal direction can be determined by the image boundary on either side, or the number of pixels of the first frame center point or the second frame center point in the horizontal direction close to each other can be determined by the image boundaries on both sides at the same time. This embodiment does not limit this. The image distance between the coordinates of the first frame center point and the coordinates of the second frame center point on the image horizontal axis can be determined by counting the number of pixels.

[0068] Among them, the distance mapping relationship is constructed based on the physical size of the monitoring area and the number of pixels in the effective image frame. Specifically, when the mobile control ball is fixed at the specified position, the actual length represented by the unit pixel can be obtained by measuring the pixel length of the target at a known distance in the image, and then dividing the actual physical length by the pixel length. The image distance in the image coordinate system can be converted into the physical distance in the physical coordinate system through the distance mapping relationship. By comparing the physical distance as the distance between the oxygen cylinder and the acetylene cylinder with the preset distance, the accuracy of the distance judgment can be guaranteed.

[0069] Based on the above description of the image distance calculation process, in order to ensure the accuracy of image distance calculation, the specific execution process for calculating the image distance between the first frame center point coordinates and the second frame center point coordinates on the horizontal axis of the image is: calculate the first pixel number and the second pixel number between the first frame center point coordinates and the second frame center point coordinates on the horizontal axis of the image and any side boundary; take the absolute value of the difference between the first pixel number and the second pixel number as the image distance.

[0070] Wherein, the boundary on either side refers to the boundary of the image on either side of the horizontal direction of the target image frame. When the boundary selected by the center point coordinates of the first frame and the center point coordinates of the second frame is the same, the image distance can be obtained by calculating the absolute value of the difference between the first pixel number and the second pixel number, and when the boundary selected by the center point coordinates of the first frame and the center point coordinates of the second frame is different, the image distance can be obtained by subtracting the first pixel number and the second pixel number from the total number of pixels on the horizontal axis of the image.

[0071] Furthermore, in order to ensure the accuracy of the calculation of the duration of absence from the job, before determining whether the duration of absence from the job supervisor is less than a preset duration based on the category information, the method also includes: detecting the number of target video frames in which the job supervisor is continuously absent in the category information; and calculating the duration of absence from the job supervisor based on the capture interval corresponding to the preset frequency and the target number of video frames.

[0072] Among them, since the target video frame is continuously intercepted at a preset frequency, when the work guardian disappears in the category information corresponding to a certain target video frame, it means that the work guardian is away from work at this time. Starting from the target video frame as the duration, the duration of the work guardian's absence can be obtained by multiplying the interception interval corresponding to the preset frequency by the number of target video frames in which the work guardian is continuously away from work. By comparing the duration of absence with the preset duration, the accuracy of duration judgment can be guaranteed.

[0073] 207. Determine whether there are illegal operations with the second hidden danger level in the hot work area, and screen out the non-compliant conditions among multiple necessary conditions.

[0074] The second hidden danger level is used to characterize safety hidden dangers.

[0075] Combined with step 205, it can be known that in this step, in order to make the subsequent alarm more targeted at the illegal operations corresponding to the preparation hidden dangers, the illegal operations can be divided into different hidden danger levels, and the second hidden danger level corresponds to the safety hidden danger. When it is detected that the post-rule is not satisfied, there is an illegal operation of the second hidden danger level. In addition, according to the judgment results of each necessary condition in the aforementioned step 206, the non-standard conditions among the multiple necessary conditions can be obtained, that is, the conditions that are not met among the necessary conditions, so that the subsequent alarm can generate corresponding violation information based on the non-standard conditions. By defining the hidden danger level and generating specific violation information, the subsequent alarm can be made more specific and accurate, that is, more targeted, so that the relevant person in charge can quickly identify the problem and take remedial measures in time.

[0076] In combination with steps 205 and 207, it should be noted that the second hidden danger level has a higher degree of urgency in the alarm operation than the first hidden danger level. For example, the first hidden danger level may trigger a text message or email notification to the relevant person in charge of the hot work, while the second hidden danger level may immediately trigger an audible and visual alarm and notify all relevant operators to stop the work.

[0077] 208. When the hidden danger level is the first hidden danger level, prepare violation information according to the generation scenario of the non-compliance category, and trigger a first alarm operation based on the prepared missing information.

[0078] In this step, detailed violation information corresponding to the pre-job preparation, that is, scene preparation violation information, can be generated according to the non-compliance category. Exemplarily, assuming that the non-compliance category is a handheld fire extinguisher, indicating that there is a lack of handheld fire extinguisher configuration in the action operation scene, the violation information may be: "There is a lack of handheld fire extinguishers in the current area". In addition, the violation information may also include hidden dangers that may be caused by the lack of the non-compliance category and recommended emergency measures, etc., which are not limited in this embodiment. The first alarm operation may specifically refer to sending alarm information to preset recipients through multiple channels such as text messages, emails, mobile application notifications, etc., including but not limited to on-site persons in charge, safety managers, and relevant regulatory agencies.

[0079] 209. When the hidden danger level is the second hidden danger level, scene safety violation information is generated according to the hidden danger degree corresponding to the non-compliance condition, and a second alarm operation is triggered based on the scene safety violation information.

[0080] In actual applications, each necessary condition has a different degree of hidden danger corresponding to it as a non-compliance condition. Among them, the safety helmet is used to protect the valve of the oxygen cylinder from damage. Once the valve is damaged, it may cause oxygen leakage or even explosion, and the shockproof ring prevents the oxygen cylinder from being damaged by collision during transportation or storage. Therefore, the hidden danger level corresponding to the oxygen cylinder without a safety helmet and / or shockproof ring can be set to primary; the distance between the oxygen cylinder and the acetylene cylinder is less than the preset distance (usually more than 5 meters). Once a leak or spark occurs, both cylinders may be ignited, resulting in a chain reaction, causing serious fire or explosion. Therefore, the hidden danger level corresponding to the distance between the oxygen cylinder and the acetylene cylinder is less than the preset distance can be set to high; the absence of the work supervisor for a period longer than the preset time may mean a lack of immediate response in an emergency, increasing the risk of the operation. Therefore, the hidden danger level corresponding to the absence of the work supervisor for a period longer than the preset time can be set to intermediate.

[0081] In this step, detailed violation information corresponding to safety during the operation can be generated according to the non-standard conditions, that is, scene safety violation information of different levels can be generated according to the degree of hidden danger. Exemplarily, assuming that the non-standard condition is that the distance between the oxygen cylinder and the acetylene cylinder is less than the preset distance, the violation information can be: "The distance between the oxygen cylinder and the acetylene cylinder is only X meters, which does not reach the Y meters specified in the safety operation regulations". In addition, the violation information can also include the hidden dangers that may be caused by the lack of the non-standard conditions and the recommended emergency measures, etc., which are not limited in this embodiment. The second alarm operation can start the sound and light alarm system on the scene, alert all the people on the scene through strong alarm sounds and flashing lights, instruct them to take emergency avoidance measures, and send alarm information to the preset recipients through multiple channels such as text messages, emails, mobile application notifications, including but not limited to the site person in charge, operation operators, safety managers and relevant regulatory agencies. Specifically, different sound and light alarm methods can be set in advance according to the degree of hidden dangers, or indicators that can reflect the degree of hidden dangers can be added to the alarm information.

[0082] Furthermore, as a response to the above Figure 1-2 The implementation of the method embodiment shown, the embodiment of the present application provides a device for detecting illegal operations in hot work scenarios based on the power plant unsafe behavior platform, the device is used to accurately detect and warn illegal operations, and reduce safety hazards in hot work scenarios. The embodiment of the device corresponds to the aforementioned method embodiment. For ease of reading, this embodiment will no longer repeat the details of the aforementioned method embodiment, but it should be clear that the device in this embodiment can correspond to all the contents of the aforementioned method embodiment. Specifically, Figure 3 As shown, the device comprises:

[0083] An acquisition unit 31 is used to acquire at least one target image frame corresponding to a hot work area, wherein the hot work area is a monitoring area where a mobile control ball is fixed at a designated position to shoot the hot work, and the target image frame is used to represent a video image frame continuously captured at a preset frequency in a video stream of the hot work area;

[0084] The processing unit 32 is used to input the at least one target image frame obtained by the acquisition unit 31 into a pre-trained target detection model to obtain category information and coordinate information of each target in the at least one target image frame;

[0085] A judging unit 33, configured to judge whether there is any illegal operation in the hot work area according to the category information and the coordinate information obtained by the processing unit 32;

[0086] The alarm unit 34 is used to perform corresponding alarm operations based on the hidden danger level of the illegal operation if the judgment unit 33 judges that there is illegal operation in the hot work area.

[0087] Further, such as Figure 4 As shown, the judging unit 33 includes:

[0088] The acquisition module 331 is used to acquire the compliance operation rules preset in the hot work area, wherein the compliance operation rules include pre-rules and post-rules, wherein the pre-rules are used to determine whether the scene preparation in the hot work area is compliant, and the post-rules are used to determine whether the scene safety in the hot work area is compliant;

[0089] A first judgment module 332 is used to judge whether the hot work area meets the pre-rule obtained by the acquisition module 331 according to the category information, and the pre-rule includes a plurality of necessary categories corresponding to the scene preparation;

[0090] A first determination module 333 is used to determine that there is a violation operation of a first hidden danger level in the hot work area if the first judgment module 332 determines that the hot work area does not meet the pre-rule, and screen out the substandard categories from the multiple necessary categories, wherein the first hidden danger level is used to characterize the preparation hidden danger;

[0091] A second judgment module 334 is used to judge whether the hot work area satisfies the post-rule according to the category information and the coordinate information if the first judgment module 332 judges that the hot work area satisfies the pre-rule, wherein the post-rule includes a plurality of necessary conditions corresponding to the scene safety;

[0092] The second determination module 335 is used to determine that there are illegal operations of the second hidden danger level in the hot work area if the second judgment module 334 determines that the hot work area does not meet the post-rules obtained by the acquisition module 331, and screen out the substandard conditions from the multiple necessary conditions. The second hidden danger level is used to characterize safety hazards.

[0093] Further, such as Figure 4 As shown, the first determination module 332 includes:

[0094] The first judgment submodule 3321 is used to judge whether there are work supervisors, handheld fire extinguishers, oxygen cylinders and acetylene cylinders in the hot work area at the same time according to the category information;

[0095] A first determination submodule 3322 is configured to determine that the hot work area satisfies the precondition rule if the first determination submodule 3321 determines that there are a work supervisor, a handheld fire extinguisher, an oxygen cylinder, and an acetylene cylinder in the hot work area at the same time;

[0096] The first determination submodule 3323 is also used to determine that the hot work area does not meet the pre-condition rule if the first judgment submodule 3321 determines that there is no work supervisor, handheld fire extinguisher, oxygen cylinder and acetylene cylinder in the hot work area.

[0097] Further, such as Figure 4 As shown, the second judgment module 334 includes:

[0098] The second judgment submodule 3341 is used to judge whether the oxygen cylinder has a safety helmet and a shockproof ring according to the category information, judge whether the duration of the work supervisor's absence from the post is less than a preset duration according to the category information, and judge whether the distance between the oxygen cylinder and the acetylene cylinder is greater than a preset distance according to the coordinate information;

[0099] The second determination submodule 3342 is used to determine that the hot work area does not meet the post-rule if the second judgment submodule 3341 determines that the oxygen cylinder does not have a safety helmet and / or a shockproof ring, determines that the duration of the work supervisor's absence from the post is greater than or equal to a preset duration, and determines that the distance between the oxygen cylinder and the acetylene cylinder is less than or equal to a preset distance;

[0100] The second determination submodule 3342 is also used to determine that the hot work area meets the post-rule if the second judgment submodule 3341 determines that the oxygen cylinder is equipped with a safety helmet and a shockproof ring, determines that the duration of the work supervisor's absence from the post is less than a preset duration, and determines that the distance between the oxygen cylinder and the acetylene cylinder is greater than a preset distance.

[0101] Further, such as Figure 4 As shown, the device also includes:

[0102] The third determination submodule 3343 is used to determine the coordinates of the center point of the first frame corresponding to the oxygen cylinder according to the frame coordinates of the oxygen cylinder before the second judgment module 3342;

[0103] The third determining submodule 3343 is further used to determine the coordinates of the center point of the second frame corresponding to the acetylene gas cylinder according to the frame coordinates of the acetylene gas cylinder;

[0104] A first calculation submodule 3344 is used to calculate the image distance between the first frame center point coordinate and the second frame center point coordinate obtained by the third determination submodule 3343 on the horizontal axis of the image;

[0105] The conversion submodule 3345 is used to convert the image distance obtained by the first calculation submodule 3344 into a physical distance by using a distance mapping relationship, and use the physical distance as the distance between the oxygen cylinder and the acetylene cylinder. The distance mapping relationship is constructed based on the physical size of the monitoring area and the number of pixels of the effective image frame.

[0106] Further, such as Figure 4 As shown, the first calculation submodule 3344 is specifically used for:

[0107] Calculate the first pixel number and the second pixel number between the first frame center point coordinate and the second frame center point coordinate on the horizontal axis of the image and any side boundary;

[0108] The absolute value of the difference between the first number of pixels and the second number of pixels is used as the image distance.

[0109] Further, such as Figure 4 As shown, the device also includes:

[0110] A detection submodule 3346 is used to detect the number of target video frames in which there is no operation supervisor in the category information continuously;

[0111] The second calculation submodule 3347 is used to calculate the duration of the absence of the work supervisor according to the interception interval corresponding to the preset frequency and the target number of video frames obtained by the detection submodule 3346.

[0112] Further, such as Figure 4 As shown, the alarm unit 34 includes:

[0113] A first alarm module 341 is used for preparing violation information according to the generation scenario of the non-compliance category when the potential danger level is the first potential danger level, and triggering a first alarm operation based on the preparation missing information;

[0114] The second alarm module 342 is used to generate scene safety violation information according to the degree of hidden danger corresponding to the non-compliance condition when the hidden danger level is the second hidden danger level, and trigger a second alarm operation based on the scene safety violation information.

[0115] Furthermore, an embodiment of the present application further provides a storage medium, wherein the storage medium is used to store a computer program, wherein when the computer program is running, the device where the storage medium is located is controlled to execute the above Figure 1-2 The method for detecting illegal operations in hot work scenarios based on the power plant unsafe behavior platform is described in.

[0116] Furthermore, the embodiment of the present application also provides a processor, the processor is used to run a program, wherein the program executes the above Figure 1-2 The method for detecting illegal operations in hot work scenarios based on the power plant unsafe behavior platform is described in.

[0117] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0118] It is understandable that the related features in the above methods and devices can be referenced to each other. In addition, the "first", "second" and the like in the above embodiments are used to distinguish the embodiments, but do not represent the advantages and disadvantages of the embodiments.

[0119] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0120] The algorithm and display provided herein are not inherently related to any particular computer, virtual system or other device. Various general purpose systems can also be used together with the teachings based on this. According to the above description, it is obvious that the structure required for constructing such systems. In addition, the application is not directed to any specific programming language either. It should be understood that various programming languages ​​can be utilized to realize the content of the application described herein, and the description of the specific language above is for the purpose of disclosing the best mode of implementation of the application.

[0121] In addition, the memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0122] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0123] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0124] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0125] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0126] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0127] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0128] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include transitory media such as modulated data signals and carrier waves.

[0129] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including 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, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0130] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0131] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A method for detecting illegal operations in hot work scenarios based on an unsafe behavior platform of a power plant, characterized in that: The method comprises: Acquire at least one target image frame corresponding to a hot work area, wherein the hot work area is a monitoring area where a mobile control ball is fixed at a designated position to shoot the hot work, and the target image frame is used to represent video image frames continuously captured at a preset frequency in a video stream of the hot work area; Inputting at least one of the target image frames into a pre-trained target detection model to obtain category information and coordinate information of each target in at least one of the target image frames; Determine whether there is any illegal operation in the hot work area according to the category information and the coordinate information; If so, a corresponding warning operation is performed based on the risk level of the illegal operation.

2. The method according to claim 1, characterized in that Judging whether there is any illegal operation in the hot work area according to the category information and the coordinate information includes: Obtaining the compliance operation rules preset for the hot work area, the compliance operation rules including pre-rules and post-rules, the pre-rules are used to determine whether the scene preparation of the hot work area is compliant, and the post-rules are used to determine whether the scene safety of the hot work area is compliant; Determining whether the hot work area satisfies the pre-conditions according to the category information, wherein the pre-conditions include a plurality of necessary categories corresponding to the scene preparation; If not, it is determined that there is illegal operation of the first hidden danger level in the hot work area, and the substandard categories in the multiple necessary categories are screened out, and the first hidden danger level is used to characterize the preparation hidden danger; If so, judging whether the hot work area satisfies the post-rule according to the category information and the coordinate information, wherein the post-rule includes a plurality of necessary conditions corresponding to the safety of the scene; If not, it is determined that there is illegal operation of the second hidden danger level in the hot work area, and the substandard conditions among the multiple necessary conditions are screened out. The second hidden danger level is used to characterize safety hazards.

3. The method according to claim 2, characterized in that Judging whether the hot work area satisfies the precondition according to the category information includes: Determine, based on the category information, whether there are work supervisors, handheld fire extinguishers, oxygen cylinders, and acetylene cylinders in the hot work area at the same time; If yes, it is determined that the hot work area satisfies the precondition rule; If not, it is determined that the hot work area does not meet the precondition rule.

4. The method according to claim 2, characterized in that: Judging whether the hot work area satisfies the post-condition according to the category information and the coordinate information includes: Determine whether the oxygen cylinder is equipped with a safety helmet and a shockproof ring according to the category information, determine whether the duration of the work supervisor's absence from work is less than a preset duration according to the category information, and determine whether the distance between the oxygen cylinder and the acetylene cylinder is greater than a preset distance according to the coordinate information; If not, it is determined that the hot work area does not satisfy the post-rule; If so, it is determined that the hot work area satisfies the post-rule.

5. The method according to claim 4, characterized in that Before determining whether the distance between the oxygen cylinder and the acetylene cylinder is greater than a preset distance according to the coordinate information, the method further includes: Determine the coordinates of the center point of the first frame corresponding to the oxygen cylinder according to the frame coordinates of the oxygen cylinder; Determine the coordinates of the center point of the second frame corresponding to the acetylene gas cylinder according to the frame coordinates of the acetylene gas cylinder; Calculating an image distance between the coordinates of the center point of the first frame and the coordinates of the center point of the second frame on the horizontal axis of the image. Specifically, calculating a first number of pixels and a second number of pixels between the coordinates of the center point of the first frame and the coordinates of the center point of the second frame and any side boundary on the horizontal axis of the image, and taking an absolute value of a difference between the first number of pixels and the second number of pixels as the image distance; The image distance is converted into a physical distance using a distance mapping relationship, and the physical distance is used as the distance between the oxygen cylinder and the acetylene cylinder. The distance mapping relationship is constructed based on the physical size of the monitoring area and the number of pixels of the effective image frame.

6. The method according to claim 4, characterized in that Before determining whether the duration of absence of the operation supervisor is less than a preset duration according to the category information, the method further includes: Detecting the number of target video frames in which there is no continuous operation supervisor in the category information; The duration of the absence of the operation supervisor is calculated according to the interception interval corresponding to the preset frequency and the target number of video frames.

7. The method according to any one of claims 2 to 6, characterized in that: Based on the risk level of the illegal operation, corresponding alarm operations are performed, including: When the hidden danger level is the first hidden danger level, preparing violation information according to the generation scenario of the non-compliance category, and triggering a first alarm operation based on the preparation missing information; When the hidden danger level is the second hidden danger level, scene safety violation information is generated according to the hidden danger degree corresponding to the non-compliance condition, and a second alarm operation is triggered based on the scene safety violation information.

8. A device for detecting illegal operations in hot work scenarios based on an unsafe behavior platform of a power plant, characterized in that: The device comprises: An acquisition unit is used to acquire at least one target image frame corresponding to a hot work area, wherein the hot work area is a monitoring area where a mobile control ball is fixed at a designated position to shoot the hot work, and the target image frame is used to represent video image frames continuously captured at a preset frequency in a video stream of the hot work area; a processing unit, configured to input the at least one target image frame obtained by the acquisition unit into a pre-trained target detection model to obtain category information and coordinate information of each target in the at least one target image frame; A judging unit, configured to judge whether there is any illegal operation in the hot work area according to the category information and the coordinate information obtained by the processing unit; The alarm unit is used to perform corresponding alarm operations based on the hidden danger level of the illegal operation if the judgment unit judges that there is illegal operation in the hot work area.

9. A storage medium, characterized in that: The storage medium includes a stored program, wherein when the program is running, the device where the storage medium is located is controlled to execute the illegal operation detection method for the hot work operation scenario based on the power plant unsafe behavior platform as described in any one of claims 1 to 7.

10. A processor, characterized in that: The processor is used to run a program, wherein when the program is run, the method for detecting illegal operations in a hot work scenario based on an unsafe behavior platform of a power plant as described in any one of claims 1 to 7 is executed.

Images