An oxyacetylene fire operation intelligent monitoring system and method

Abstract

The application discloses an oxyacetylene torch operation intelligent monitoring system and method, which comprises the following steps: 1, real-time image acquisition of the monitoring area is carried out through an information acquisition subprogram; then, data preprocessing is carried out through a data processing subprogram, and a target detection algorithm YoloV5 is used to identify and locate the target in the image; 2, the relationship between the specific target relative positions in the visible light image and the far infrared image is detected through a flame detection subprogram, a high-temperature area without visible fire or a flame area is judged and analyzed on site, and oxyacetylene flame and fire flame are distinguished; 3, the real-time state of the monitoring area is fed back through a result output subprogram, and the monitoring result is output. The application improves the monitoring accuracy, does not need supervision, and improves the safety of the torch operation.

Description

Technical Field

[0001] This invention relates to the field of image processing technology for oxyacetylene hot work safety monitoring, and particularly to an intelligent monitoring system and method for oxyacetylene hot work operations. Background Technology

[0002] Hot work refers to temporary work that may generate sparks, flames, and hot surfaces. Common hot work includes electric welding, gas welding (thermal cutting), electric drilling, grinding wheels, and blowtorches. However, if hot work is performed improperly, accidents and fires can easily occur.

[0003] With the increasingly widespread application of oxyacetylene flames in industrial production, the safe and efficient use of these flames has become a growing concern. Oxyacetylene flames can easily ignite other flammable materials in the working environment, potentially leading to fires if not detected promptly. Fires are a serious threat to public safety and people's lives, often causing significant casualties and property damage. Therefore, fire monitoring in fire-prone areas is essential.

[0004] According to relevant regulations, one or two safety officers must be present at the hot work site to supervise the operation. However, these officers leave the site immediately after the operation, making continuous monitoring impossible. High-temperature spots or smoldering areas may remain undetected immediately after the operation, posing potential safety hazards after personnel leave. Therefore, existing hot work monitoring methods are labor-intensive and cannot provide continuous monitoring. Summary of the Invention

[0005] In order to overcome the shortcomings of the prior art, the purpose of this invention is to provide an intelligent monitoring system and method for oxyacetylene hot work, which improves the accuracy of monitoring, eliminates the need for supervision, and enhances the safety of hot work operations.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0007] An intelligent monitoring system for oxyacetylene hot work includes an information acquisition subroutine, a data processing subroutine, a flame detection subroutine, and a result output subroutine.

[0008] The information acquisition subroutine uses a visible light sensor and a far-infrared sensor to collect information from the area to be monitored, and obtains visible light color images and far-infrared images.

[0009] The data processing subroutine performs digital image processing on the acquired information, removing some noise generated during the acquisition process through Gaussian filtering, color enhancement, and erosion and dilation of the image to obtain clear image information.

[0010] The flame detection subroutine extracts and identifies the feature information of the target by recognizing clear image information, and determines the flame type;

[0011] The result output subroutine provides feedback on the real-time status of the monitoring area determined by the flame detection subroutine and outputs the monitoring results.

[0012] A method for intelligent monitoring of oxyacetylene hot work operations includes the following steps;

[0013] Step 1: The information acquisition subroutine acquires images of the area to be monitored in real time; then the data processing subroutine performs preprocessing; and finally, the YOLOv5 target detection algorithm is used to identify and locate the targets in the images.

[0014] Step 2: Based on the collected information, the flame detection subroutine analyzes the relationship between the relative positions of specific targets in the visible light and far-infrared images, identifies high-temperature areas without open flames or flame areas, and distinguishes between oxyacetylene flames and fire flames.

[0015] Step 3: Feedback on the real-time status of the monitored area through the result output subroutine and output the monitoring results.

[0016] In step 1, the acquired information is preprocessed digitally. Through Gaussian filtering, color enhancement, and erosion and dilation of the image, noise generated during the acquisition process is removed to obtain clear image information.

[0017] The target detection algorithm YOLOv5 identifies and locates targets in an image in the following ways:

[0018] By extracting and identifying the feature information of the target to be identified, the specific coordinate information of the flame, welding torch and operator in the monitoring area is obtained. The extracted category and specific coordinate information are compared with the temperature information of the corresponding position in the far-infrared image to verify the authenticity of the flame. Then, it is compared with the specific position of the welding torch to determine the flame category. Finally, the safety of the operator is determined by the operator's activity trajectory, and the presence of danger in the monitoring area is confirmed.

[0019] Determining whether there are safety hazards in the work environment mainly involves two situations: target detection through visible light images. If flames are detected, it indicates that there is an open flame at the work site or that the fire is in its early stages. Otherwise, it indicates that there is no open flame but there is smoldering or continuous high temperature.

[0020] The characteristic information of the identified target includes the flame, the welding torch, and the operator.

[0021] In step 2, the specific steps for distinguishing between oxyacetylene flames and fire flames are as follows:

[0022] Step 1: Determine if a high-temperature area exists in the detected flame. Obtain the specific coordinates of the flame from the visible light image by monitoring the area. The enclosed area is the flame's location. Then, read the specific coordinates of the high-temperature area from the far-infrared image. If the area has high temperature, compare the coordinates of the two images. If the coordinates of the flame in the visible light image and the coordinates of the high-temperature area in the far-infrared image coincide or there are more than 20 overlapping pixels, it proves that the detected flame is a real flame. Otherwise, it may be a false detection caused by interference from factors such as lighting during the target detection process, and the conclusion that the working environment is safe can be directly drawn.

[0023] Step 2: Determine if the flame is an oxyacetylene flame. After confirming the authenticity of the flame, determine whether the flame is an oxyacetylene flame by using the coordinate information of the flame and the welding torch. If the coordinate information of the flame and the coordinate information of the welding torch overlap by more than 20 pixels or the minimum coordinate difference between the two is less than 30 pixels, that is, the flame and the welding torch are physically connected, or the flame and the welding torch do not have a visual connection due to the excessive pressure of the gas ejected by the welding torch.

[0024] A threshold range for the distance between the two must be set. If the above conditions are met, it proves that the flame is an oxyacetylene flame. If the distance between the flame and the welding torch is too far, that is, the coordinate information of the flame and the welding torch is greater than the set threshold range, then the flame is a fire flame, that is, a safety accident has occurred in the hot work area, and it needs to be dealt with in time.

[0025] Step 3: Determine if the hot work site is safe. After confirming that the flame is an oxyacetylene flame, it is necessary to determine if the hot work site is safe. If there are no workers operating at the site, or if a worker has an accident and falls to the ground motionless while the oxyacetylene flame is still burning, it will also cause a safety accident and start a fire. In scenarios where an oxyacetylene flame is present, workers must be operating the flame. The distance between the worker and the flame will be very small or there will be some overlap. Furthermore, the worker will definitely have some movement during the operation and will not remain motionless. Therefore, after obtaining the coordinate information of the oxyacetylene flame and the worker, if there is an overlapping area and the worker's coordinates keep changing, it proves that the monitored area is safe. If there is no worker information in the monitored area or the worker's coordinates remain unchanged, it proves that there is a safety hazard in the monitored area, and an early warning message needs to be issued in time.

[0026] In step 2, the step of determining whether the monitoring area is safe when there is no open flame in the monitoring area is as follows: if there is a high temperature area in the monitoring area but no open flame, it is necessary to determine whether there is a safety hazard in this area.

[0027] There are only two situations that can cause high temperatures in an area: hot work has just been carried out in the area, and the object that has just been worked on will definitely have a high temperature area, or there may be smoldering of combustibles in the area, which will cause high temperatures. Therefore, if the distance difference between the coordinates of the high temperature area and the coordinates of the last operation is greater than 80 pixels, it proves that there may be smoldering in the high temperature area, which needs to be dealt with in time and an early warning should be issued.

[0028] If the coordinates of the high-temperature area coincide with the coordinates of the previous operation, it indicates that this area needs to be continuously monitored to determine whether the object just finished in this area is still hot or there is still smoldering of flammable materials. If the far-infrared information shows that the temperature in this area is gradually decreasing, it indicates that the operation has just ended and that there is no smoldering. If the dimensions of this area remain unchanged within the set time, there is a safety hazard here, and an early warning should be issued in time.

[0029] The beneficial effects of this invention are:

[0030] This invention enables real-time monitoring of safety at oxyacetylene hot work sites. Compared to traditional methods involving safety personnel, it saves significant manpower costs and allows for continuous monitoring, preventing accidents caused by safety personnel negligence. Traditional video fire monitoring detectors often produce false alarms when monitoring hot work scenarios, causing the system to malfunction. The method proposed in this invention can effectively solve this problem. Attached Figure Description

[0031] Figure 1 This is the main flowchart of the system of the present invention.

[0032] Figure 2 This is a system flowchart for flame detection according to the present invention.

[0033] Figure 3 This is a flowchart illustrating the monitoring process for fire status during information fusion in this invention.

[0034] Figure 4 This is a flowchart illustrating the monitoring process for the fireless state during information fusion in this invention. Detailed Implementation

[0035] The present invention will now be described in further detail with reference to the accompanying drawings.

[0036] like Figure 1 As shown: The intelligent monitoring method for oxy-acetylene hot work operations utilizing machine vision and multi-source information fusion. The main flowchart of the system of this invention includes the following steps:

[0037] After the system is powered on, the four subroutines "information acquisition subroutine", "data processing subroutine", "flame detection subroutine" and "result output subroutine" start running. Through the operation of the above subroutines, the intelligent monitoring of the hot work site is finally completed.

[0038] In the information acquisition process, this invention mainly uses visible light sensors and far-infrared sensors to collect information about the area to be monitored, thereby obtaining visible light color images and far-infrared images.

[0039] The data processing subroutine mainly performs digital image processing on the acquired information. Through Gaussian filtering, color enhancement, and image erosion and dilation, some noise generated during the acquisition process is removed, and finally clear image information is obtained.

[0040] like Figure 2 As shown: The flame detection procedure of the present invention has the following steps:

[0041] After obtaining a clear image from the data processing subroutine, the YOLOv5 target detection algorithm is used to extract and identify the feature information of the target in the image that needs to be clearly identified. Finally, the specific coordinate information of the flame, welding torch, and operator in the monitoring area is obtained. Then, the extracted category and specific coordinate information are compared with the temperature information of the corresponding position in the far-infrared image to obtain the authenticity of the flame. Then, it is compared with the specific position of the welding torch to determine the flame category. Finally, the operator's activity trajectory is used to determine whether the operator is safe, and finally, it is confirmed whether there is any danger in the monitoring area.

[0042] Determining whether there are safety hazards in the work environment mainly involves two situations: target detection through visible light images. If flames are detected, it indicates that there is an open flame at the work site or that the fire is in its early stages. Otherwise, it indicates that there is no open flame but there is smoldering or continuous high temperature.

[0043] like Figure 3 As shown: The specific process of determining whether a flame is an oxyacetylene flame in this invention is as follows:

[0044] Step 1: Determine if a high-temperature region exists within the detected flame. The specific coordinates of the flame are obtained from the visible light image using a target detection algorithm. For example, the area enclosed by (10,10), (40,40), (10,40), and (40,10) represents the flame's location. Then, the specific coordinates of the high-temperature region are read from the far-infrared image. For example, the area containing (10,20), (10,20), (30,20), and (30,10) represents a high-temperature region. The coordinates of the two images are compared. If the coordinates of the flame in the visible light image coincide with the coordinates of the high-temperature region in the far-infrared image, or if there is more than 20 overlapping pixels, it proves that the detected flame is a real flame. Otherwise, it may be a false detection caused by interference from factors such as lighting, leading to the conclusion that the working environment is safe.

[0045] Step 2: Determine if the flame is an oxyacetylene flame. After confirming the authenticity of the flame, determine whether it is an oxyacetylene flame by using the coordinate information of the flame and the welding torch. If the coordinate information of the flame and the welding torch overlaps by more than 20 pixels or the minimum coordinate difference between them is less than 30 pixels, it means that the flame and the welding torch are physically connected. Alternatively, if the pressure of the gas ejected from the welding torch is too high, causing the flame and the welding torch to not be visually connected, a distance threshold range must be set between them. If the above conditions are met, it proves that the flame is an oxyacetylene flame. If the distance between the flame and the welding torch is too far, that is, the coordinate information of the flame and the welding torch exceeds the set threshold range, then the flame is a fire flame, indicating a safety accident in the hot work area, which needs to be dealt with immediately.

[0046] Step 3: Determine the safety of the hot work site. After confirming that the flame is an oxyacetylene flame, it is necessary to determine whether the hot work site is safe. If there are no workers operating at the site, or if a worker has an accident and falls to the ground motionless while the oxyacetylene flame continues to burn, it will also cause a safety accident and start a fire. In scenarios where an oxyacetylene flame is present, workers will inevitably be required to operate the flame. The distance between the worker and the flame will be very small or there will be some overlap. Furthermore, the worker will definitely be moving around during the operation and will not remain motionless. Therefore, after obtaining the coordinate information of the oxyacetylene flame and the worker, if there is an overlapping area and the worker's coordinates are constantly changing, it proves that the monitored area is safe. If there is no worker information in the monitored area or the worker's coordinates remain unchanged, it proves that there is a safety hazard in the monitored area, and an early warning message needs to be issued in time.

[0047] like Figure 4 The flowchart illustrates the process of determining the safety of a monitoring area when there is no open flame within the monitoring area, as per the present invention.

[0048] If a high-temperature area exists within the monitoring zone but no open flame is detected, it is necessary to determine whether this area poses a safety hazard. There are only two possibilities for a high temperature in an area: either a hot work operation has just been performed, and the object immediately handled will inevitably have a high-temperature zone, or there may be smoldering flammable material in the area, causing the high temperature. Therefore, if the distance difference between the coordinates of the high-temperature area and the coordinates of the previous operation is greater than 80 pixels, it indicates that there may be smoldering in this high-temperature area, requiring immediate action and an early warning should be issued. If the coordinates of the high-temperature area coincide with the coordinates of the previous operation, it indicates that this area needs continuous monitoring to determine whether the recently handled object is hot or there is smoldering flammable material. If the far-infrared information shows that the temperature in this area is gradually decreasing, it indicates that the operation has just ended, rather than smoldering. If the dimensions of this area remain unchanged within the set time, then there is a safety hazard, and an early warning should be issued promptly.

[0049] In the result output program of this invention, the monitoring device itself and the wristband worn by the monitor can issue audible and visual alarms. The background monitoring device can also issue alarms and perform operations such as resetting the system. The monitoring history and alarm logs can be queried in the background software.

[0050] The system uses backend software to monitor images of the hot work site in real time. If smoldering or fire occurs, the system can notify monitoring personnel to take timely measures through various alarm devices such as alarms on the monitoring equipment, wristbands worn by monitoring personnel, and mobile apps.

Claims

1. A method for using an intelligent monitoring system for oxyacetylene hot work operations, characterized in that, The method comprises the following steps: Step 1: real-time image acquisition of the monitoring area by the information acquisition subprogram; then pre-processing by the data processing subprogram, and then target detection algorithm YoloV5 is used to identify and locate the target in the image; Step 2: for the collected information, the relationship between the specific target relative positions in the visible light image and the far infrared image is determined by the flame detection subprogram, and the high-temperature area without visible fire or the flame area is judged and analyzed on site, and the oxyacetylene flame and the fire flame are distinguished; Step 3: the real-time state of the monitoring area is fed back by the result output subprogram, and the monitoring result is output; In step 2, the specific steps for distinguishing the oxyacetylene flame and the fire flame are as follows: First step: determine whether the detected flame has a high-temperature area, obtain the specific coordinate information of the flame from the visible light image of the monitoring area, and the area surrounded by the flame is the flame area, then read the specific coordinates of the high-temperature area from the far infrared image, and the area exists high temperature, compare the coordinates of the two, when the coordinates of the flame in the visible light image coincide with the coordinates of the high-temperature area in the far infrared image or there are more than 20 pixel points overlapping area, it is proved that the detected flame is a real flame, otherwise it is a false detection in the target detection process caused by the interference of light factors, and a conclusion is directly drawn that the working environment is safe; Second step: determine whether the flame is an oxyacetylene flame, after determining the authenticity of the flame, determine whether the flame is an oxyacetylene flame through the coordinate information of the flame and the welding gun, when the coordinate information of the flame and the coordinate information of the welding gun exist more than 20 pixel points overlap or the minimum coordinate difference of the two is less than 30 pixel points, that is, the flame and the welding gun are connected in the physical layer, or the flame and the welding gun are not connected in the visual layer due to the excessive gas pressure of the welding gun; A threshold range of distance is set between the two, if the above conditions are met, it is proved that the flame is an oxyacetylene flame; if the distance between the flame and the welding gun is too far, that is, the coordinate information of the flame and the welding gun is greater than the set threshold range, then the flame is a fire flame, that is, a safety accident occurs in the hot work area, which needs to be handled in time; Third step: determine whether the hot work site is safe, after determining that the flame is an oxyacetylene flame, it is necessary to determine whether the hot work site is safe, if there is no worker operating in the work site, or the worker has an accident and is motionless, and the oxyacetylene flame is still burning, it will also cause a safety accident and trigger a fire, in the scene where the oxyacetylene flame appears, the worker must operate, the distance between the worker and the flame must be small or there must be some overlap, and the worker must have a certain amount of activity during the operation and cannot remain motionless, therefore, after obtaining the coordinate information of the oxyacetylene flame and the worker, if there is an overlapping area between the two, and the worker coordinate information changes all the time, it is proved that the monitoring area is safe, if there is no worker information in the monitoring area or the worker coordinate information remains unchanged, it is proved that there is a safety hazard in the monitoring area, and warning information needs to be sent in time. The method is realized by an oxyacetylene fire operation intelligent monitoring system, which comprises an information acquisition subprogram, a data processing subprogram, a flame detection subprogram and a result output subprogram; The information acquisition subprogram uses a visible light sensor and a far infrared sensor to collect information of the monitored area, and obtains a visible light color image and a far infrared image; The data processing subprogram performs digital image processing on the collected information, removes some noise generated in the collection process through Gaussian filtering, color enhancement and erosion and expansion of the image, and obtains clear image information; The flame detection subprogram identifies the clear image information, extracts and identifies the feature information of the target, and judges the flame category; The result output subprogram feeds back the real-time state of the monitored area judged by the flame detection subprogram, and outputs the monitoring result.

2. The method of using an oxyacetylene fire operation intelligent monitoring system according to claim 1, characterized in that, The target detection algorithm YoloV5 identifies and locates the target in the image, specifically: By extracting and identifying the feature information of the target to be identified, the specific coordinate information of the flame, the welding gun and the operator in the monitored area is obtained, the extracted category and specific coordinate information are compared with the temperature information of the corresponding position in the far infrared image, the authenticity of the flame is verified, then the specific position of the welding gun is compared, the flame category is judged, and whether the operator is safe is determined through the activity track of the operator, and finally it is determined whether there is danger in the monitored area.

3. The method of using an oxyacetylene fire operation intelligent monitoring system according to claim 2, wherein, Determining whether there is a safety hazard in the working environment mainly includes two cases: Through target detection of the visible light image, if the flame is monitored, there is an open fire in the working site or in the early stage of the fire; Otherwise, there is no open fire but smoldering or sustained high temperature.

4. The method of using an oxyacetylene fire operation intelligent monitoring system according to claim 3, wherein, In step 2, if there is no open fire in the monitored area, the step of determining whether the monitored area is safe is: if there is a high temperature area in the monitored area but no open fire, it is necessary to determine whether there is a safety hazard in this area; There are only two cases that cause high temperature in the area: this area has just undergone a fire operation, and the object just finished the operation must have a high temperature area, or there is smoldering of combustible materials in this area, causing high temperature, so if the distance between the coordinates of the high temperature area and the coordinates of the last operation is greater than 80 pixel points, it proves that there is smoldering in the high temperature area, which needs to be handled in time, and a warning information should be issued; If the coordinates of the high temperature area and the coordinate information of the last operation coincide, it proves that this area needs to be paid attention to, and it is determined whether the high temperature of the object just finished the operation or the smoldering of combustible materials exists, in this area, if the far infrared information shows that the temperature of this area gradually decreases, it proves that this place is just finished the operation, not smoldering, if the temperature of this area remains unchanged within a set time, there is a safety hazard in this area, and a warning information should be issued in time.

Images