Mine electric spark recognition and alarm method based on image immediate frame pixel features

Abstract

The present invention discloses a mine electric spark recognition and alarm method based on image pre-frame pixel features. The electric spark recognition and alarm method collects and identifies suspected electric spark images in the monitoring area in real time by installing a high-definition camera underground, and performs mine electric spark recognition and alarm by monitoring the electric spark image and pre-frame image features and the electric spark interference source image and pre-frame image features. The electric spark recognition and alarm method fully considers the characteristics of electric spark release and image, and has fast, accurate and simple recognition, which can buy precious rescue and escape time for people in distress underground.

Description

Technical Field

[0001] The invention relates to a mine electric spark recognition and alarm method based on image immediate frame pixel characteristics, and the method relates to the fields of digital image processing technology and communication technology. Background Art

[0002] Electric sparks in mines (hereinafter referred to as electric sparks) can cause gas and coal dust explosions. Gas and coal dust explosions can also trigger fires, causing secondary damage, which increases the difficulty of emergency rescue after the disaster. The mutual conversion of fire and explosion has caused the disaster area to increase rapidly, the danger level to increase suddenly, and lead to a vicious cycle of disasters. Therefore, it is crucial to quickly identify underground electric sparks, promptly report to the police, and launch emergency plans and emergency rescue in a targeted manner to save the lives of people in distress underground and to ensure safe coal production.

[0003] When sparks are generated, they instantly release strong light, and the sound of discharge is relatively small. The sparks generated include sparks generated by mine cables and electrical equipment due to service life, environment, maintenance, improper repair, etc.; due to high-power radio transmission, electromotive force is induced on the metal support and electromechanical equipment underground. When the induced electromotive force is high, the discharge to the ground generates sparks with smaller discharge energy. The temperature of the spark itself is high, but the temperature rise of objects around the spark is not obvious. Mine sparks, especially mine sparks with smaller discharge energy, are difficult to monitor. So far, there is no effective method for monitoring mine sparks.

[0004] The spark release cycle can be roughly divided into four stages: dielectric breakdown, spark generation, spark maintenance and spark extinction; research data show that in the simulated spark circuit, whether it is a resistive circuit, an inductive circuit or a capacitive circuit, the spark duration period is in the microsecond level. Among them, some researchers have obtained the spark release period of about 1.5ms based on a high-voltage spark generator and a high-speed camera. Secondly, in the industrial application field, based on the characteristics of electric sparks, the release period of electric sparks during the processing of mechanical parts without damage to parts is about 0.3μs to 4ms. Research data show that the release duration of electric sparks is extremely short; and the shortest flashing frequency period of the flashing interference source included in the underground electric spark interference source is about 1s, where the light-dark ratio is about 150ms bright and about 850ms dark. In summary, the light duration period of the electric spark interference source, whether it is a flashing interference source or a constantly bright interference source, is much longer than the light duration period of the electric spark. Therefore, the present invention proposes to use the image frame pixel characteristics to identify and alarm mine electric sparks. The method is accurate, fast and simple. Summary of the invention

[0005] The purpose of the present invention is to provide a mine spark recognition and alarm method based on image immediate frame pixel features, which method fully considers the mine spark image and immediate frame image features and the spark interference source image and immediate frame image features. The spark release moment will produce a local strong light and after the monitoring video image with sparks is processed by framing within the set frame rate threshold range, the frame image with sparks has the characteristics of a single frame without continuous frames, and the spark interference source exists in at least two consecutive frames or more within the set frame rate threshold range. Therefore, the method of using the image immediate frame pixel features to judge whether there are sparks in the monitoring area can quickly identify sparks and spark interference sources, is accurate and simple, and can buy more rescue and escape time for people in distress underground.

[0006] The electric spark identification and alarm method includes that when the electric spark is generated in the mine, the light duration is extremely short; the electric spark interference source has a long light duration, and the electric spark identification and alarm in the mine are performed by using the difference between the pixel characteristics of the electric spark image and the adjacent frame image and the pixel characteristics of the electric spark interference source image and the adjacent frame image; the identification and alarm method includes the following steps:

[0007] A1: Install the spark monitoring and alarm system in the areas with cables, electromechanical and mechanical equipment, tunnels, coal mining faces, belts and excavation faces, and initialize the system;

[0008] A2: In the electric spark monitoring alarm system, multiple image monitoring devices collect video images of the monitoring area in real time, and process the monitoring video images by frame according to the set condition 1;

[0009] A3: Real-time diagnosis of monitoring video images collected by multiple image monitoring devices in the electric spark monitoring alarm system; according to the set condition 2, determine whether there is a suspected electric spark abnormal frame image in the monitoring video image after framing;

[0010] A4: loop A3, when it is determined that a suspected spark abnormal frame image appears in the monitoring video image, the spark monitoring alarm device sends out an image abnormality warning signal;

[0011] A5: When an image abnormality warning signal appears, when it is determined that the pixel characteristic values ​​of the suspected spark abnormality frame image and the pixel characteristic values ​​of the two adjacent frame images respectively meet the set condition 3, execute step A6, otherwise execute step A3;

[0012] A6: When it is determined that the pixel characteristic values ​​of the suspected spark abnormal frame image and the pixel characteristic values ​​of the two adjacent frame images respectively meet the setting condition 4, execute step A7, otherwise execute step A3;

[0013] A7: If the current abnormal frame image is determined to be a real spark image, the spark monitoring and alarm device sends a mine spark alarm signal to the monitoring terminal above the well.

[0014] 1. The spark recognition and alarm method further includes: the setting condition 1 is that the frame rate FR of the monitoring video image is within the set threshold value (r1, r2), and the threshold value r1 is based on the longest light duration D of the spark. Tmax Set and The threshold value r2 is based on the shortest light duration G of the electric spark interference source. Tmin Set and The bright duration D of the spark and spark interference source Tmax , G Tmin Determined by experiments or based on artificial settings.

[0015] 2. The electric spark identification and alarm method further includes: the setting condition 2 for determining the suspected electric spark abnormal frame image is that the absolute value of the difference between the sum of single-channel pixel feature values ​​of any two adjacent frame images in the monitoring video image after framing is greater than the set pixel threshold N1; the any two adjacent frame images are continuous r frame images intercepted in the monitoring video image after framing, and the sum of single-channel pixel feature values ​​of the intercepted single frame images are respectively calculated; further, three consecutive middle frame images are taken from the r frame images and set as the front frame image, the middle frame image, and the rear frame image; then the two adjacent frame images include the front frame image and the middle frame image, and the middle frame image and the rear frame image; the number of frames r includes at least five frames and the number of frames r is an odd number.

[0016] 3. The electric spark identification and alarm method further includes: the determination that the pixel characteristic values ​​of the suspected electric spark abnormal frame image and the pixel characteristic values ​​of the two adjacent frame images respectively meet the set condition 3; the set condition 3 is that the difference between the suspected electric spark abnormal frame image and the corresponding continuous single-point pixel characteristic values ​​on the two previous and next frame images is greater than the set pixel threshold N2 and the number of the differences between the suspected electric spark abnormal frame image and the corresponding continuous single-point pixel characteristic values ​​on the two adjacent frame images that are greater than the threshold N2 is less than the threshold Q; the thresholds N2 and Q are determined by experiments or set manually.

[0017] 4. The electric spark identification and alarm method further includes: the judgment that the pixel characteristic values ​​of the suspected electric spark abnormal frame image and the pixel characteristic values ​​of the two adjacent frame images respectively meet the setting condition 4; the setting condition 4 is that the sum of the differences between the suspected electric spark abnormal frame image and the continuous single-point pixel characteristic values ​​corresponding to the previous and next two frames of images is greater than K times the absolute value of the difference between the single-channel pixel characteristic values ​​of the previous and next two frames of images respectively, and the ratio of the absolute value of the difference between the suspected electric spark abnormal frame image and the single-channel pixel characteristic values ​​of the previous and next two frames of images is within the threshold (A, B); the threshold K∈(k1, k2) and the threshold (A, B) are obtained by experimental measurement or artificial setting.

[0018] 5. The spark identification and alarm method further includes: the mine sparks include sparks generated by resistive circuits, sparks generated by inductive circuits and sparks generated by capacitive circuits; the spark light duration is the continuous lighting time when the spark is generated; the spark interference source light duration is the continuous lighting time of the spark interference source; the underground spark interference source includes a flickering interference source and a long-bright interference source; including car lights, mining lamps, emergency lights, tunnel lights, signal lights, reflective objects, mechanical or electromechanical equipment status indicator lights.

[0019] 6. The electric spark identification and alarm method further includes: the electric spark monitoring and alarm system includes a communication network, an electric spark monitoring and alarm device, a multi-channel image monitoring device and a monitoring terminal; the electric spark monitoring and alarm device is installed on the underground tunnel wall, tunnel roof or a fixed object used for a long time, and is directly connected to the multi-channel image monitoring device, i.e., a camera, and is responsible for receiving and processing the monitoring video image data collected by the camera. It has built-in graphic image processing software and has explosion-proof, video image data storage, timing and alarm functions; the image monitoring device includes a high-definition visible light camera, a high-definition near-infrared camera and a high-definition ultraviolet camera; the monitoring terminal is installed above the well, used to receive the alarm signal of the electric spark monitoring and alarm device, and has sound and light alarm functions.

[0020] The mine electric spark recognition and alarm method based on image immediate frame pixel features has the following characteristics:

[0021] 1. The spark recognition method of the present invention fully analyzes the characteristics of the spark release process image. Under the set image framing frame rate, the spark image exists on a single frame image and is not continuous, while under this frame rate, there are at least two consecutive frames or more of the spark interference source image. Therefore, the method of the present invention that uses the characteristic relationship of the pixels of the previous and next frames of the image to determine whether there is a spark in the monitoring area can quickly identify the spark and the spark interference source, which is accurate and simple, and can buy more rescue and escape time for the people in distress underground.

[0022] 2. The image processing algorithm of the method of the present invention is concise and efficient, and the spark information can be quickly determined by a single frame image. The feature extraction of the method has high recognition and few variables, thereby shortening the time for distinguishing suspected sparks, and further buying more escape time for trapped and affected personnel underground. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 Schematic diagram of the implementation plan of the mine electric spark recognition and alarm method based on the image frame pixel characteristics.

[0024] Figure 2 Schematic diagram of the structure of electric spark monitoring and alarm equipment.

[0025] Figure 3 Schematic diagram of the spark alarm determination process of the mine spark recognition and alarm method based on the image immediate frame pixel characteristics. DETAILED DESCRIPTION

[0026] Figure 1 This is an example of an implementation plan for a mine electric spark recognition and alarm method based on image frame pixel features, and its main components include:

[0027] 1. A storage server (101), wherein the storage server is connected in communication with an image acquisition camera (105) and is responsible for storing and forwarding real-time video image data of a monitoring area provided by the camera; the storage server is connected in communication with an electric spark monitoring alarm device (106) and is responsible for storing and forwarding electric spark alarm data provided by the electric spark monitoring alarm device; and provides a monitoring terminal (102) with a service for viewing on-site monitoring data of a monitored area or retrieving historical monitoring image data.

[0028] 2. A monitoring terminal (102) is installed on the well and is used to receive the alarm signal of the spark monitoring alarm device (106); it is responsible for providing underground environment monitoring data display services, and the storage server (101) provides real-time, historical data and spark alarm data, and has sound and light alarm functions; production management personnel can retrieve and query the historical data stored in the storage server (101) through the monitoring terminal.

[0029] 3. Core switch (103), the core management and switching device of mining Ethernet, is responsible for the management and data exchange of all devices connected to mining Ethernet, has routing function, and is connected to the Internet.

[0030] 4. Ring network switch (104), underground switching equipment of mining Ethernet, installed underground, multiple ring network switches are connected in a ring network manner.

[0031] 5. Image monitoring equipment (105), i.e., cameras, are installed in areas where electric sparks are likely to occur, such as cables, electromechanical equipment, mechanical equipment, tunnels, coal mining working faces, belts and excavation working faces, to collect monitoring video images in real time; the cameras include high-definition visible light cameras, high-definition near-infrared cameras and high-definition ultraviolet cameras; the cameras have network output and analog video output functions; the network interface is directly connected to the ring network switch (104) to transmit the real-time collected video image data of the monitoring area to the storage server (101), and the analog video output port is connected to the electric spark monitoring alarm device (106).

[0032] 6. The spark monitoring alarm device (106) is responsible for receiving and processing the monitoring video image data collected by the camera (105), and has built-in graphic image video processing software. When the processed abnormal frame image meets the set spark alarm condition, it sends the spark alarm data to the storage server (101); and sends the spark alarm signal to the monitoring terminal (102); it has wired and wireless communication functions.

[0033] 7. Suspected spark area: The camera monitors the suspected spark area in real time. The image data is monitored and collected and uploaded by the camera in real time.

[0034] Figure 2 It is a schematic diagram of the structure of an electric spark monitoring and alarm device, and the structure of the electric spark monitoring and alarm device mainly includes:

[0035] Core processor, graphic image processor, storage unit, clock unit, power supply unit, USB interface unit, video image acquisition module, wireless communication unit, network interface unit, SD card interface unit.

[0036] 1. Core processor (201), using Broadcom BCM2837 processor, using ARM Cortex-A53 architecture, 64-bit quad-core 1.2GHz.

[0037] 2. Graphics image processor (202), which adopts Dual Core VideoCore IV GPU processor.

[0038] 3. Storage unit (203), using 1GB LPDD2 memory.

[0039] 4. Clock unit (204), using 19.2MHz crystal oscillator.

[0040] 5. The power supply unit (205) adopts an AC / DC module, with an input of 100V~240VAC and an output of 12VDC, which is used to power the equipment.

[0041] 6. USB interface unit (206), supporting 4 USB interfaces.

[0042] 7. The video image acquisition module (207) converts the analog video signal into digital video data, inputs the analog video output port connected to the camera (105), transmits the digital video data to the core processor (201) through the USB port, and supports multi-channel video acquisition.

[0043] 8. Communication module (208), responsible for converting the RS485 communication interface into a USB communication interface to connect other data monitoring devices.

[0044] 9. A wireless communication unit (209), supporting 802.11b / g / n protocol, and used for wireless communication with a monitoring device supporting wireless communication.

[0045] 10. The network interface unit (210) is responsible for accessing the mining Ethernet and connecting to the ring network switch (104).

[0046] 11. SD card unit (211), used to store system files, library files, monitoring program files, etc., using Linux system management, built-in OpenCV library for video data processing, using a Micro SD card of no less than 4GB.

[0047] An example of the spark alarm determination process of the mine spark recognition and alarm method based on the image frame pixel characteristics is as follows: Figure 3 As shown, the process includes:

[0048] 1. (301) The camera is installed in areas with cables, electromechanical and mechanical equipment, tunnels, coal mining faces, belts and excavation faces, etc., to collect video images of the monitored area in real time and upload them to the electric spark monitoring and alarm equipment for image recognition and frame processing, and further execute step (302).

[0049] 2. (302) The spark monitoring alarm device pre-processes the monitoring video image, performs frame division operation on the monitoring video image based on the frame rate (r1, r2) and obtains n frames of M×N images (a1, a2···a n-1 ,a n )(n∈N + ), execute step (303).

[0050] 3. (303) intercepting continuous r (r∈n) frames of images from the video image after frame processing, and further calculating the single channel pixel feature values ​​and Said is a frame image (a1, a2···a r ) and further execute step (304).

[0051] 4. (304) The further step is to select three consecutive frames of images in the middle of the r frames to perform pixel feature value and relationship determination on adjacent frames, and execute step (305).

[0052] 5. (305) When two adjacent frame images The absolute value of the difference between the single channel pixel feature value and the absolute value of S1 = |A a[((r+1) / 2)-1] -A a((r+1) / 2) |>N1 or when two adjacent frame images The absolute value of the difference between the single channel pixel feature value and the absolute value of S2 = |A a[((r+1) / 2)+1] -Aa((r+1) / 2) |>N1, the threshold N1 is obtained by experimental measurement or artificial setting; the A a[((r+1) / 2)-1] ,A a((r+1) / 2) ,A a[((r+1) / 2)+1] For frame image The single channel pixel feature value and; execute step (306), otherwise return to execute step (303).

[0053] 6. (306) Determine if an abnormal frame image appears in the video image and the electric spark monitoring alarm device sends a video image abnormality warning signal to the monitoring terminal, and execute step (307).

[0054] 7. (307) The specific frame value of the abnormal frame image is further determined, and when the condition is met When , the abnormal frame image is determined to be Frame; execute step (308). Otherwise, execute step (310)

[0055] 8. (308) When two adjacent frame images The difference of the corresponding continuous single-point pixel feature values and two adjacent frame images The difference of the corresponding continuous single-point pixel feature values And frame image With two adjacent frame images The above correspond to the number of consecutive single-point pixel feature values ​​whose difference is greater than the threshold N2. ; execute step (309); otherwise, return to execute step (303).

[0056] 9. (309) When two adjacent frame images The sum of the differences of the corresponding continuous single-point pixel feature values and two adjacent frame images The sum of the differences of the corresponding continuous single-point pixel feature values And when S2 / S1∈(A,B), execute step (318); otherwise, return to execute step (303).

[0057] 10. (310) The current frame image The single channel pixel feature value and A a[((r+1) / 2)-1] With the ad-frame image The corresponding pixel feature value and A a((r+1) / 2) Satisfy condition A a[((r+1) / 2)-1] >A a((r+1) / 2) When , the abnormal frame image is frame; further execute step (311), otherwise execute step (314).

[0058] 11. (311) When the abnormal frame image is determined to be a frame image When the frame image is set is the middle frame image; further determine the frame image If the pixel feature value relationship is found, execute step (312), otherwise return to execute step (303).

[0059] 12. (312) When two adjacent frame images The difference of the corresponding continuous single-point pixel feature values and two adjacent frame images The difference of the corresponding continuous single-point pixel feature values And frame image With two adjacent frame images The above correspond to the number of consecutive single-point pixel feature values ​​whose difference is greater than the threshold N2. , execute step (313); otherwise, return to execute step (303).

[0060] 13. (313) When two adjacent frame images The sum of the differences of the corresponding continuous single-point pixel feature values and two adjacent frame images The sum of the differences of the corresponding continuous single-point pixel feature values And S3 / S1∈(A,B), the threshold S3 is the current frame image The absolute value of the difference between the single-channel pixel feature value and , execute step (318), otherwise return to execute step (303).

[0061] 14.(314) The current frame image The single channel pixel feature value and A a[((r+1) / 2)+1] With two adjacent frame images The corresponding pixel feature value and A a((r+1) / 2) Satisfy condition A a[((r+1) / 2)+1] >A a((r+1) / 2) When , the abnormal frame image is Frame; further execute step (315), otherwise return to execute step (303).

[0062] 15. (315) When the abnormal frame image is determined to be a frame image When the frame image is set is the middle frame image; further determine the frame image If the pixel feature value relationship is not satisfied, execute step (316); otherwise, return to execute step (303).

[0063] 16.(316) When two adjacent frame images The difference of the corresponding continuous single-point pixel feature values and two adjacent frame images The difference of the corresponding continuous single-point pixel feature values And frame image With two adjacent frame images The above correspond to the number of consecutive single-point pixel feature values ​​whose difference is greater than the threshold N2. , execute step (317); otherwise, return to execute step (303).

[0064] 17.(317) When two adjacent frame images The sum of the differences of the corresponding continuous single-point pixel feature values and two adjacent frame images The sum of the differences of the corresponding continuous single-point pixel feature values When S4 / S2∈(A,B), the threshold S4 is the value of the current frame image. The absolute value of the difference between the single-channel pixel feature value and , execute step (318), otherwise return to execute step (303).

[0065] 18.(318) Send an electric spark alarm signal to the monitoring terminal.

Claims

1. A mine electric spark recognition and alarm method based on image immediate frame pixel features, characterized in that: When a mine electric spark is generated, the duration of the light is extremely short; the duration of the light of the electric spark interference source is long. The difference between the pixel characteristics of the electric spark image and the immediate frame image and the pixel characteristics of the electric spark interference source image and the immediate frame image is used to identify and alarm the mine electric spark; the identification and alarm method includes the following steps: A1: Install the spark monitoring and alarm system in the areas with cables, electromechanical and mechanical equipment, tunnels, coal mining faces, belts and excavation faces, and initialize the system; A2: In the electric spark monitoring alarm system, multiple image monitoring devices collect video images of the monitoring area in real time, and process the monitoring video images by frame according to the set condition 1; A3: Real-time diagnosis of monitoring video images collected by multiple image monitoring devices in the electric spark monitoring alarm system; according to the set condition 2, determine whether there is a suspected electric spark abnormal frame image in the monitoring video image after framing; A4: loop A3, when it is determined that a suspected spark abnormal frame image appears in the monitoring video image, the spark monitoring alarm device sends out an image abnormality warning signal; A5: When an image abnormality warning signal appears, when it is determined that the pixel characteristic values ​​of the suspected spark abnormality frame image and the pixel characteristic values ​​of the two adjacent frame images respectively meet the set condition 3, execute step A6, otherwise execute step A3; A6: When it is determined that the pixel characteristic values ​​of the suspected spark abnormal frame image and the pixel characteristic values ​​of the two adjacent frame images respectively meet the setting condition 4, execute step A7, otherwise execute step A3; A7: If the current abnormal frame image is determined to be a real spark image, the spark monitoring and alarm device sends a mine spark alarm signal to the monitoring terminal above the well.

2. The mine electric spark recognition and alarm method based on image immediate frame pixel features as claimed in claim 1, characterized in that: The setting condition 1 is that the frame rate FR of the monitoring video image is within the set threshold value (r1, r2), and the threshold value r1 is based on the longest light duration D of the electric spark. Tmax Set and The threshold value r2 is based on the shortest light duration G of the electric spark interference source. Tmin Set and The bright duration D of the spark and spark interference source Tmax , G Tmin Determined by experiments or based on artificial settings.

3. The mine electric spark recognition and alarm method based on image immediate frame pixel features as claimed in claim 1, characterized in that: The setting condition 2 for determining a suspected electric spark abnormal frame image is that the absolute value of the difference between the sum of single-channel pixel feature values ​​of any two adjacent frame images in the monitoring video image after framing is greater than the set pixel threshold N1; the any two adjacent frame images are continuous r frame images intercepted in the monitoring video image after framing, and the sum of single-channel pixel feature values ​​of the intercepted single frame images is calculated respectively; further, three consecutive middle frame images are taken from the r frame images and set as the front frame image, the middle frame image, and the rear frame image; then the two adjacent frame images include the front frame image and the middle frame image, and the middle frame image and the rear frame image; the number of frames r includes at least five frame images and the number of frames r is an odd number.

4. The mine electric spark recognition and alarm method based on image immediate frame pixel features as claimed in claim 1, characterized in that: The determination is that the pixel characteristic values ​​of the suspected electric spark abnormal frame image and the pixel characteristic values ​​of the two adjacent frame images respectively meet the set condition 3; the set condition 3 is that the difference between the suspected electric spark abnormal frame image and the corresponding continuous single-point pixel characteristic values ​​on the two previous and next frame images is greater than the set pixel threshold N2 and the number of differences between the suspected electric spark abnormal frame image and the corresponding continuous single-point pixel characteristic values ​​on the two adjacent frame images that are greater than the threshold N2 is less than the threshold Q; the thresholds N2 and Q are determined by experiments or set manually.

5. The mine electric spark recognition and alarm method based on image immediate frame pixel features as claimed in claim 1, characterized in that: The pixel characteristic values ​​of the suspected electric spark abnormal frame image and the pixel characteristic values ​​of the two adjacent frame images respectively meet the setting condition 4; the setting condition 4 is that the sum of the differences between the suspected electric spark abnormal frame image and the continuous single-point pixel characteristic values ​​corresponding to the previous and next two frames of images is greater than K times the absolute value of the difference between the single-channel pixel characteristic values ​​of the previous and next two frames of images respectively, and the ratio of the absolute value of the difference between the suspected electric spark abnormal frame image and the single-channel pixel characteristic values ​​of the previous and next two frames of images is within the threshold (A, B); the threshold K∈(k1, k2) and the threshold (A, B) are obtained by experimental measurement or artificial setting.

6. The mine electric spark recognition and alarm method based on image immediate frame pixel features as claimed in claim 1, characterized in that: The mine sparks include sparks generated by resistive circuits, sparks generated by inductive circuits and sparks generated by capacitive circuits; the spark light duration is the continuous lighting time when the spark is generated; the spark interference source light duration is the continuous lighting time of the spark interference source; underground spark interference sources include flickering interference sources and long-bright interference sources, including car lights, mining lamps, emergency lights, tunnel lights, signal lights, reflective objects, and mechanical or electromechanical equipment status indicator lights.

7. The mine electric spark recognition and alarm method based on image immediate frame pixel features as claimed in claim 1, characterized in that: The spark monitoring and alarm system includes a communication network, spark monitoring and alarm equipment, multi-channel image monitoring equipment and a monitoring terminal; the spark monitoring and alarm equipment is installed on the underground tunnel wall, tunnel roof or a fixed object used for a long time, and is directly connected to the multi-channel image monitoring equipment, i.e., the camera, and is responsible for receiving and processing the monitoring video image data collected by the camera. It has built-in graphic image processing software and has explosion-proof, video image data storage, timing and alarm functions; the image monitoring equipment includes a high-definition visible light camera, a high-definition near-infrared camera and a high-definition ultraviolet camera; the monitoring terminal is installed above the well, used to receive the alarm signal of the spark monitoring and alarm equipment, and has sound and light alarm functions.

Images