Mine fire identification and alarm method based on image contour equal-area perimeter ratio

Abstract

The application discloses a mine fire identification and alarm method based on image contour equal-area perimeter ratio, which comprises the following steps: installing a camera in a mine, collecting and identifying a suspected fire image of a monitoring area in real time, and identifying and alarming a mine fire by monitoring irregular shape characteristics of flame combustion. The mine fire identification and alarm method fully considers the image characteristics of fire combustion, is fast, accurate, simple and easy to identify, and can help people in distress in the mine to gain valuable rescue and escape time.

Description

Technical Field

[0001] This invention relates to a method for identifying and alarming mine fires based on the perimeter ratio of equal areas of image contours. This method involves fields such as digital image processing technology and communication technology. Background Technology

[0002] In the coal industry, accidents such as gas leaks, fires, floods, roof collapses, and coal dust plague safe coal mine production. Once a mine fire breaks out, if the fire cannot be controlled in time, its spread will rapidly expand, causing numerous casualties and property damage; furthermore, it may act as a ignition source, causing explosions in areas with high concentrations of gas and coal dust underground, resulting in secondary damage to the underground environment. Therefore, quickly identifying underground fire hazards, promptly reporting them, and activating targeted emergency plans and rescue operations are crucial for safe coal mining production.

[0003] Existing mine fire monitoring technologies include various sensor-based integrated monitoring methods, such as temperature, smoke, and gas sensors. While these methods offer advantages like high reliability and ease of operation, their deployment and maintenance are labor-intensive due to the complex underground working environment, and they are highly susceptible to interference. Infrared radiation thermometry can measure the temperature of fire sources, but its accuracy is significantly affected by underground environmental factors and fire interference sources. Visual feature monitoring is currently the mainstream mine fire monitoring technology, but existing technologies rely on numerous characteristic variables for fire assessment. While these features are rich in information, they also contain many unstable factors. Assessing fire conditions based on multiple cameras and frames leads to longer assessment times and higher rates of missed and false alarms.

[0004] Therefore, it is necessary to research new methods for rapid identification of mine fires, shorten the judgment time, and reduce the false alarm and false negative rates of existing monitoring technologies for mine fire detection. Early detection of mine fires and the ability to issue fire warnings as soon as possible are crucial for timely emergency rescue and saving the lives of people trapped underground. Summary of the Invention

[0005] The purpose of this invention is to provide a method for identifying and alarming mine fires based on the perimeter ratio of the area of ​​an image contour. Since flames have irregular shapes, and interfering light sources include, but are not limited to, fluorescent lamps, incandescent lamps, LED lamps, mine lamps, and vehicle lights, the method fully considers the characteristics of fire combustion images, enabling rapid identification of fires and fire interference sources. It is accurate and simple, and can buy more rescue and escape time for personnel in distress underground. The method involves installing cameras in roadways, chambers, tunneling faces, and fully mechanized mining faces with cables, conveyor belts, and electromechanical equipment to monitor the monitored area in real time. Fire alarms are determined based on whether the ratios of the perimeter of the circle with the area of ​​the suspected fire area on the suspected fire image to its actual perimeter, the perimeter of the triangle with the area of ​​the suspected fire area to its actual perimeter, and the perimeter of the rectangle with the area of ​​the suspected fire area to its actual perimeter meet set conditions. Fires in the monitored underground area are also determined based on whether the ratio of the number of fire alarms determined within a set time to the total number of determinations meets set conditions. When a fire is determined to occur in the monitored underground area, a fire alarm signal is immediately sent to the monitoring terminal.

[0006] The fire alarm determination process includes:

[0007] Step 1: The camera acquires images of the monitored area in real time. After image filtering and noise reduction, image enhancement, and image pixel binarization, the monitored images are used to identify single-area pixel feature values ​​in the image. Greater than the set threshold When the image is determined to be a suspected fire image of a single area; when the image contains multiple disconnected areas of pixel feature values... All are greater than the set threshold. If the image is determined to be a suspected fire image in multiple areas, step 2 is executed; if the suspected fire image determination criteria are not met, monitoring continues.

[0008] Step 2: If there is only a single suspected fire area in the suspected fire image, directly process the contour edge features of the suspected fire area in the suspected fire image to calculate the actual area and actual perimeter of the suspected fire area contour; or if there are multiple suspected fire areas in the suspected fire image, segment the multiple suspected fire areas, process the contour edge features of each segment, calculate the actual area and actual perimeter of the suspected fire area contour, and then proceed to Step 3.

[0009] Step 3: Calculate the perimeter of the circle with the same area as the suspected fire area on the suspected fire image and calculate the ratio of its perimeter to the actual perimeter of its outline; calculate the perimeter of the triangle with the same area as the suspected fire area and calculate the ratio of its perimeter to the actual perimeter of its outline; calculate the perimeter of the rectangle with the same area as the suspected fire area and calculate the ratio of its perimeter to the actual perimeter of its outline, then proceed to Step 4.

[0010] Step 4: When the ratios of the perimeter of a circle with the same area as the suspected fire area to its actual outline perimeter, the ratio of the perimeter of a triangle with the same area as the suspected fire area to its actual outline perimeter, and the ratio of the perimeter of a rectangle with the same area as the suspected fire area to its actual outline perimeter in the suspected fire image meet the following conditions: When, in the formula These are, respectively, the ratio of the perimeter of a circle with an area equal to that of a suspected fire region in the suspected fire image to its actual outline perimeter; the ratio of the perimeter of a triangle with an area equal to that of a suspected fire region to its actual outline perimeter; and the ratio of the perimeter of a rectangle with an area equal to that of a suspected fire region to its actual outline perimeter. To set a threshold; the If the result is obtained through experimental determination or manual setting, proceed to step 5; otherwise, return to step 1.

[0011] Step 5: Activate the fire alarm and return to step 1.

[0012] 1. The fire identification method further includes: the fire determination process in the underground monitoring area includes a duration of... The process of determining a fire alarm is performed in a cyclical manner, and the number of suspected fire image alarms that meet the fire alarm determination criteria is recorded. Total number of judgments for the image The ratio is greater than the set threshold. At that time, that is If so, it is determined that a fire has occurred in the underground monitoring area.

[0013] 2. The fire identification method further includes: the circumference of the circle representing the suspected fire area on the suspected fire image is obtained based on the formula for the actual area and circumference of the suspected fire area. In the formula The actual area of ​​the suspected fire region on the suspected fire image; the ratio of the circumference of the circle representing the suspected fire region on the suspected fire image to its actual outline circumference: In the formula The perimeter of the suspected fire area on the suspected fire image is given.

[0014] 3. The fire identification method further includes: the perimeter of the triangle with equal area of ​​the suspected fire area on the suspected fire image is obtained based on the formula for the actual area of ​​the suspected fire area and the perimeter of the triangle. In the formula The actual area of ​​the suspected fire zone on the suspected fire image. The ratio of the lengths of any two sides to the third side of a triangle with an area equal to that of the suspected fire area in an image of a suspected fire; the ratio The perimeter of the smallest circumscribed triangle of the suspected fire area is equivalently obtained by the ratio of the lengths of any two sides to the third side; the ratio of the perimeter of the equal-area triangle of the suspected fire area on the suspected fire image to its actual perimeter is: In the formula The perimeter of the suspected fire area on the suspected fire image is given.

[0015] 4. The fire identification method further includes: the perimeter of the rectangle with an area equal to the suspected fire area on the suspected fire image is obtained based on the formula for the actual area of ​​the suspected fire area and the perimeter of the rectangle. In the formula The actual area of ​​the suspected fire zone on the suspected fire image. The ratio of the side lengths of rectangles with equal areas representing the suspected fire area in a suspected fire image; the ratio The ratio of any side of the smallest bounding rectangle of the suspected fire area to its adjacent side is equivalently obtained; the ratio of the perimeter of the rectangle with the same area of ​​the suspected fire area on the suspected fire image to its actual perimeter is: In the formula The perimeter of the suspected fire area on the suspected fire image is defined as the actual outline of the area; the shape with the equal perimeter is not limited to circles, triangles, and rectangles.

[0016] The mine fire identification and alarm method based on the number of edge corners of the image contour has the following characteristics:

[0017] 1. The fire identification method of the present invention fully analyzes the contour features of fire combustion images and the image contour features of underground fire interference sources. The method determines the fire based on the ratio of the perimeter of a circle, the perimeter of a triangle, and the perimeter of a rectangle with the same area to the actual contour perimeter of the suspected fire area on the suspected fire image. This is more conducive to identifying and eliminating fire interference sources and can reduce false alarms and missed alarms of fire disasters.

[0018] 2. The method of the present invention not only makes efficient use of the feature information of fire images, but also has a simple and efficient algorithm for image processing. Fire information can be quickly determined from a single frame image. The method has high feature extraction recognition and few variables, thereby shortening the identification time of suspected fires and further giving trapped and affected personnel more time to escape. Attached Figure Description

[0019] Figure 1 A schematic diagram of the implementation scheme for a mine fire identification and alarm method based on the perimeter ratio of equal area of ​​image contour.

[0020] Figure 2 Schematic diagram of fire monitoring and alarm equipment.

[0021] Figure 3 A schematic diagram of the fire alarm process for a mine fire identification and alarm method based on the perimeter ratio of equal area of ​​image contour.

[0022] Figure 4 A schematic diagram of the fire determination process for a mine fire identification and alarm method based on the perimeter ratio of equal area of ​​image contour. Detailed Implementation

[0023] An example implementation plan for a mine fire identification and alarm method based on the equal area perimeter ratio of image contours is shown below. Figure 1 As shown, the main components include:

[0024] 1. A storage server (101) is communicatively connected to an image acquisition camera (105) and is responsible for storing and forwarding real-time video image data of the monitoring area provided by the camera; the storage server is communicatively connected to a fire monitoring and alarm device (106) and is responsible for storing and forwarding fire alarm data provided by the fire monitoring and alarm device; and provides a service for the monitoring terminal (102) to view the on-site monitoring data of the area to be monitored or to retrieve historical monitoring image data.

[0025] 2. Monitoring terminal (102), installed on the surface of the well, is used to receive alarm signals from the fire monitoring and alarm equipment (106); it is responsible for providing underground environmental monitoring data display services, and the storage server (101) provides real-time, historical data and fire alarm data, and has audible and visual alarm functions; production management personnel can retrieve and query the historical data stored in the storage server (101) through the monitoring terminal.

[0026] 3. Core Switch (103): The core management and switching device for mining Ethernet, responsible for the management and data exchange of all devices connected to the mining Ethernet, with routing function and connection to the Internet.

[0027] 4. Ring network switch (104), an underground switching device for mining Ethernet, installed underground, with multiple ring network switches connected in a ring network manner.

[0028] 5. Camera (105), an image acquisition device, installed in underground roadways, coal mining faces and tunneling faces, responsible for acquiring real-time video images of areas prone to fire, such as underground roadways and working faces; the camera includes a visible light camera, a far-infrared camera, a near-infrared camera and an ultraviolet camera; the camera has network output and analog video output functions; the network interface is directly connected to a ring network switch (104) to transmit the real-time acquired video image data of the monitoring area to a storage server (101), and the analog video output port is connected to a fire monitoring and alarm device (106).

[0029] 6. Fire monitoring and alarm equipment (106) is responsible for receiving and processing monitoring video image data collected by camera (105). It has built-in graphics and image processing software. When the processed suspected fire image meets the set fire alarm conditions, it sends fire alarm data to storage server (101) and sends fire alarm signal to monitoring terminal (102). It has wired and wireless communication functions.

[0030] 7. Suspected fire area: Areas suspected of being on fire that appear in the camera's real-time monitoring area. This image data is monitored, collected, and uploaded by the camera in real time.

[0031] Figure 2 This is a schematic diagram of the fire monitoring and alarm equipment:

[0032] The fire monitoring and alarm equipment mainly consists of: a core processor, a graphics processor, a storage unit, a clock unit, a power supply unit, a USB interface unit, a video image acquisition module, a wireless communication unit, a network interface unit, and an SD card interface unit.

[0033] 1. Core processor (201), using Broadcom BCM2837 processor, adopting ARM Cortex-A53 architecture, 64-bit quad-core 1.2GHz.

[0034] 2. Graphics processor (202), which is a Dual Core VideoCore IV GPU processor.

[0035] 3. Storage unit (203) uses 1GB LPDD2 memory.

[0036] 4. Clock unit (204), using a 19.2MHz crystal oscillator.

[0037] 5. Power supply unit (205), adopts AC / DC module, input 100V~240VAC, output 12VDC, used for power supply of equipment.

[0038] 6. USB interface unit (206) supports 4 USB interfaces.

[0039] 7. The video image acquisition module (207) converts the modal video signal into digital video data, inputs it to the analog video output port of the camera (105), and transmits the digital video data to the core processor (201) via the USB port, supporting multi-channel video acquisition.

[0040] 8. Communication module (208) is responsible for converting the RS485 communication interface into a USB communication interface to connect other data monitoring devices.

[0041] 9. Wireless communication unit (209) supporting the 802.11b / g / n protocol for wireless communication with monitoring equipment that supports wireless communication.

[0042] 10. Network interface unit (210) is responsible for accessing the mining Ethernet and connecting to the ring network switch (104).

[0043] 11. SD card unit (211) is used to store system files, library files, monitoring program files, etc. It is managed by Linux system, has a built-in OpenCV library for video data processing, and uses a Micro SD card of not less than 4GB.

[0044] An example of a fire alarm process for a mine fire identification and alarm method based on the equal area perimeter ratio of image contours is shown below. Figure 3 As shown, the process includes:

[0045] 1. (301) Set the internal timer of the fire monitoring and alarm equipment to a fixed time. Set the value to zero and start the timer, while simultaneously setting the variable representing the total number of times the images that meet the fire alarm judgment criteria will be evaluated. Set a variable for the number of fire alarms that meet the fire detection alarm criteria based on suspected fire images. Execute step (302).

[0046] 2. (302) Further set the variable for the total number of judgments of the image that executes the fire alarm judgment condition. Each time the monitoring image executes the fire alarm judgment condition, the judgment count variable is incremented by one, and step (303) is executed.

[0047] 3. (303) The fire monitoring and alarm equipment processes the image and makes a fire alarm judgment on the suspected fire image. When the set fire alarm judgment conditions are met, step (304) is executed; otherwise, step (302) is returned to be executed.

[0048] 4. (304) For each suspected fire image that meets the fire alarm determination condition, the fire alarm count variable is incremented by one, i.e. At the same time, a fire warning is sent to the monitoring terminal, and step (305) is executed.

[0049] 5. (305) The fire alarm determination process is executed cyclically within a set time period to determine the timer value. Is it greater than the set time threshold? When the time value If the condition is met, proceed to step (306); otherwise, return to step (302).

[0050] 6. (306) Further determine the number of fire image alarms that meet the fire alarm determination conditions. Total number of judgments of images that meet the fire alarm judgment criteria Is the ratio greater than the set threshold? ,when If the condition is met, proceed to step (307); otherwise, return to step (301).

[0051] 7. (307) The fire monitoring and alarm equipment sends a fire alarm signal to the storage server (101) and the monitoring terminal (102).

[0052] An example of the fire determination process for a mine fire identification and alarm method based on the equal area perimeter ratio of image contours is as follows: Figure 4 As shown, the process includes:

[0053] 1. (401) The camera is installed in the roadway, chamber and fully mechanized mining face with cables, tape and electromechanical equipment to collect images of the monitored area in real time and upload them to the fire monitoring and alarm equipment for image recognition processing, and then further execute step (402).

[0054] 2. (402) The fire monitoring and alarm equipment preprocesses the monitoring image. After image denoising, image enhancement, and image pixel binarization, the monitoring image has a single area pixel brightness feature value. The image is a suspected fire image of a single area; when there are multiple disconnected areas in the image, the pixel brightness feature values... If the image is a suspected fire image of multiple areas, proceed to step (403); otherwise return to step (401).

[0055] 3. (403) Further determine whether there are multiple suspected fire areas on the suspected fire image. If there are no multiple suspected fire areas on the suspected fire image, then determine that there is only a single suspected fire area and execute step (404); otherwise, execute step (405).

[0056] 4. (404) Based on step (403), if there is only a single suspected fire area on the suspected fire image, then the suspected fire area on the suspected fire image is directly processed for contour edge features to further calculate the actual area of ​​the suspected fire area and the actual perimeter of the suspected fire area contour; the calculation of the actual area of ​​the suspected fire area contour is based on the formula In the formula The coordinates of the vertices on the contour; the actual perimeter of the contour is calculated using the formula. In the formula Given the set of coordinates of points on the contour, execute step (407).

[0057] 5. (405) Based on step (403), if there are multiple suspected fire areas on the suspected fire image, the suspected fire areas on the suspected fire image are divided into blocks and step (406) is executed.

[0058] 6. (406) Calculate the actual area and perimeter of the suspected fire area in blocks. The calculation method is based on the formula in step (404). Execute step (407).

[0059] 7. (407) Equivalently represent the actual area of ​​the suspected fire area on the suspected fire image as the area of ​​a circle, and calculate the circumference of the circle with the same area of ​​the suspected fire area based on the formula for the actual area of ​​the outline and the circumference of the circle. Further, the ratio of the circumference of the circle with the suspected fire area on the suspected fire image to its actual outline circumference is calculated: The actual area of ​​the suspected fire area in the suspected fire image is equivalent to the area of ​​a triangle, and the perimeter of the triangle with the same area as the suspected fire area is calculated based on the formula for the actual area of ​​the outline and the perimeter of the triangle. In the formula The calculation formula is In the formula Given the minimum circumscribed triangle of the suspected fire area, further calculate the ratio of the perimeter of the equal-area triangle of the suspected fire area on the suspected fire image to its actual perimeter. The actual area of ​​the suspected fire region in the suspected fire image is equivalent to the area of ​​a rectangle, and the perimeter of the rectangle equivalent to the area of ​​the suspected fire region is calculated based on the formula for the actual area of ​​the outline and the perimeter of the rectangle. In the formula The calculation formula is In the formula Given the side lengths of the two adjacent sides of the smallest bounding rectangle of the suspected fire area, further calculate the ratio of the perimeter of the rectangle with the same area of ​​the suspected fire area on the suspected fire image to its actual perimeter: Execute step (408).

[0060] 4. (408) When the ratio of the circumference of the circle with equal area of ​​the suspected fire area on the suspected fire image to its actual outline circumference meets the set conditions. If the condition is met, proceed to step (409); otherwise, return to step (401).

[0061] 5. (409) When the ratio of the perimeter of the triangle with equal area of ​​the suspected fire area on the suspected fire image to the perimeter of its actual outline meets the set conditions. If the above steps are not executed, proceed to step (410); otherwise, return to step (401).

[0062] 6. (410) When the ratio of the perimeter of the rectangle with the same area of ​​the suspected fire area on the suspected fire image to its actual outline perimeter meets the set conditions. If the previous step is executed, proceed to step (411); otherwise, return to step (401).

[0063] 7. (411) Conduct fire alarm processing.

Claims

1. A method for mine fire identification and alarm based on the equal area perimeter ratio of image contours, characterized in that: Cameras are installed in roadways, chambers, tunneling faces, and fully mechanized mining faces with cables, conveyor belts, and electromechanical equipment to monitor the monitored area in real time. Fire alarms are determined based on whether the ratio of the perimeter of the circle with the area of ​​the suspected fire area on the suspected fire image to its actual perimeter, the ratio of the perimeter of the triangle with the area of ​​the suspected fire area to its actual perimeter, and the ratio of the perimeter of the rectangle with the area of ​​the suspected fire area to its actual perimeter meet the set conditions. Fires in the underground monitoring area are determined based on whether the ratio of the number of times a fire alarm is determined to occur to the total number of determinations within a set time meets the set conditions. When a fire is determined to occur in the underground monitoring area, a fire alarm signal is immediately sent to the monitoring terminal. The fire alarm determination process includes: Step 1: The camera acquires images of the monitored area in real time. After image filtering and noise reduction, image enhancement, and image pixel binarization, the monitored images are used to identify single-area pixel feature values ​​in the image. Greater than the set threshold When the image is determined to be a suspected fire image of a single area; when the image contains multiple disconnected areas of pixel feature values... All are greater than the set threshold. If the image is determined to be a suspected fire image in multiple areas, step 2 is executed; if the suspected fire image determination criteria are not met, monitoring continues. Step 2: If there is only a single suspected fire area in the suspected fire image, directly process the contour edge features of the suspected fire area in the suspected fire image to calculate the actual area and actual perimeter of the suspected fire area contour; or if there are multiple suspected fire areas in the suspected fire image, segment the multiple suspected fire areas, process the contour edge features of each segment, calculate the actual area and actual perimeter of the suspected fire area contour, and then proceed to Step 3. Step 3: Calculate the perimeter of the circle with the same area as the suspected fire area on the suspected fire image and calculate the ratio of its perimeter to the actual perimeter of its outline; calculate the perimeter of the triangle with the same area as the suspected fire area and calculate the ratio of its perimeter to the actual perimeter of its outline; calculate the perimeter of the rectangle with the same area as the suspected fire area and calculate the ratio of its perimeter to the actual perimeter of its outline, then proceed to Step 4. Step 4: When the ratios of the perimeter of a circle with the same area as the suspected fire area to its actual outline perimeter, the ratio of the perimeter of a triangle with the same area as the suspected fire area to its actual outline perimeter, and the ratio of the perimeter of a rectangle with the same area as the suspected fire area to its actual outline perimeter in the suspected fire image meet the following conditions: When, in the formula These are, respectively, the ratio of the perimeter of a circle with an area equal to that of a suspected fire region in the suspected fire image to its actual outline perimeter; the ratio of the perimeter of a triangle with an area equal to that of a suspected fire region to its actual outline perimeter; and the ratio of the perimeter of a rectangle with an area equal to that of a suspected fire region to its actual outline perimeter. To set a threshold; the If the result is obtained through experimental determination or manual setting, proceed to step 5; otherwise, return to step 1. Step 5: Activate the fire alarm and return to step 1.

2. The mine fire identification and alarm method based on the equal area perimeter ratio of image contours as described in claim 1, characterized in that: The fire detection process in the underground monitoring area includes a duration of... The process of performing fire alarm determination is repeated every second. The variable is the number of suspected fire image alarms that meet the fire alarm determination criteria. Total number of judgments for the image The ratio is greater than the set threshold. At that time, that is If so, it is determined that a fire has occurred in the underground monitoring area.

3. The mine fire identification and alarm method based on the equal area perimeter ratio of image contours as described in claim 1, characterized in that: The circumference of the circle representing the suspected fire area on the suspected fire image is calculated using the formula for the actual area of ​​the suspected fire area and the circumference of the circle: In the formula The actual area of ​​the suspected fire region on the suspected fire image; the ratio of the circumference of the circle representing the suspected fire region on the suspected fire image to its actual outline circumference: In the formula The perimeter of the suspected fire area on the suspected fire image is given.

4. The mine fire identification and alarm method based on the equal area perimeter ratio of image contours as described in claim 1, characterized in that: The perimeter of the triangle with equal area in the suspected fire area on the suspected fire image is obtained based on the formula for the actual area of ​​the suspected fire area and the perimeter of the triangle: In the formula The actual area of ​​the suspected fire zone on the suspected fire image. The ratio of the lengths of any two sides to the third side of a triangle with an area equal to that of the suspected fire area in an image of a suspected fire; the ratio The perimeter of the smallest circumscribed triangle of the suspected fire area is equivalently obtained by the ratio of the lengths of any two sides to the third side; the ratio of the perimeter of the equal-area triangle of the suspected fire area on the suspected fire image to its actual perimeter is: In the formula The perimeter of the suspected fire area on the suspected fire image is given.

5. The mine fire identification and alarm method based on the equal area perimeter ratio of image contours as described in claim 1, characterized in that: The perimeter of the rectangle representing the suspected fire area in the suspected fire image is calculated based on the actual area of ​​the suspected fire area and the formula for the perimeter of the rectangle: In the formula The actual area of ​​the suspected fire zone on the suspected fire image. The ratio of the side lengths of rectangles with equal areas representing the suspected fire area in a suspected fire image; the ratio The ratio of any side of the smallest bounding rectangle of the suspected fire area to its adjacent side is equivalently obtained; the ratio of the perimeter of the rectangle with the same area of ​​the suspected fire area on the suspected fire image to its actual perimeter is: In the formula The perimeter of the suspected fire area on the suspected fire image is given.

Images