Coal mine underground video dust filtration monitoring method based on model fusion

Abstract

The invention provides a coal mine underground video dust filtration monitoring method based on model fusion, and belongs to the technical field of coal mine underground video dust filtration monitoring. The technical problem to be solved is to provide the coal mine underground video dust filtration monitoring method based on model fusion. According to the technical scheme, video data of an underground operation area environment are collected in real time through a camera installed in an underground coal mine, the video data are uploaded to a server arranged in a roadway in real time, the server firstly intercepts image frames from the underground coal mine video, timestamp marking is conducted on the image frames, and the image frames are sent to the roadway; carrying out histogram equalization preprocessing on each image frame, and improving the brightness and contrast of the collected video image to realize image enhancement; the server further adopts a dark channel prior method to carry out data recovery on the video image, a fuzzy effect caused by dust on the image is removed, and the real definition of the image is recovered; the method is applied to coal mine underground video dust filtration monitoring.

Description

Technical Field

[0001] The invention provides a coal mine underground video dust filtration monitoring method based on model fusion, belonging to the technical field of coal mine underground video dust filtration monitoring. Background Art

[0002] Roadheaders are important equipment for tunnel excavation in mines. The excavation efficiency of roadheaders is a key factor in ensuring the connection between mining and excavation in coal mines, which puts higher requirements on the excavation efficiency and safety of roadheaders. At present, when carrying out excavation operations, it is still necessary to arrange an assistant driver on site to observe the excavation situation and cooperate with the main driver to complete the operation. This method has low excavation efficiency and requires the assistant driver to always observe the situation on site to ensure the normal progress of the operation. His personal safety will be threatened by many factors.

[0003] Although it is currently possible to increase video surveillance so that personnel can monitor and operate in an offline manner, the working environment underground in coal mines is complex and dangerous, mainly characterized by low light intensity, dust and narrow working space. Existing underground coal mine monitoring systems usually rely on cameras installed in different locations to capture video images of the working area. The quality of underground video images is often severely reduced due to dust obstruction, low light and jitter caused by equipment movement, which directly affects the effectiveness and accuracy of monitoring.

[0004] At present, the method for dust filtering in underground coal mine videos is mainly implemented by image dehazing technology. The main technical means include image enhancement-based method, image restoration-based method and fusion-based method, but the effect of this method is limited when the dust concentration is high; the dust filtering method based on image restoration can also reverse the influence of dust on the image through physical model or statistical model, so as to restore a clear image, but the calculation complexity is high and it is difficult to apply in real time; or the dust filtering method based on fusion combines the advantages of multiple dust filtering technologies to provide a more robust dust filtering effect, but it is difficult to implement. Video dust filtering requires dust filtering for each frame of the video. Therefore, how to provide an efficient and robust coal mine underground video dust filtering method in combination with the corresponding image dust filtering technology while ensuring real-time performance has become a problem that needs to be solved urgently. Summary of the invention

[0005] In order to overcome the deficiencies in the prior art, the present invention aims to solve the technical problem of providing a coal mine underground video dust filtration monitoring method based on model fusion.

[0006] In order to solve the above technical problems, the technical solution adopted by the present invention is: a coal mine underground video dust filtration monitoring method based on model fusion, comprising the following monitoring steps: Step 1: The video data of the underground working area environment is collected in real time through the camera installed in the coal mine, and the video data is uploaded to the server set up in the tunnel in real time. The server first captures the image frame from the underground video of the coal mine, and timestamps the image frame, and then pre-processes each image frame with histogram equalization to improve the brightness and contrast of the collected video image to achieve image enhancement; Step 2: The server further uses a dark channel prior method to recover the video image data, remove the blurring effect caused by dust on the image, and restore the true clarity of the image; Step 3: The server processes the image enhancement and image restoration steps in parallel by adopting a multi-threaded pipeline processing method; Step 4: The server uses a weighted image addition algorithm to fuse the results of image enhancement and image restoration to combine the advantages of both, generate multiple final images with optimized brightness, contrast, and clarity, and connect the final images into a video for output; Step 5: The server sends the output video to the base station controller on the tunnel boring machine through the wireless network, and the tunnel boring machine driver views it in real time.

[0007] When performing the image enhancement operation, the step 1 specifically calculates the grayscale histogram and cumulative distribution function of the image, maps the grayscale value of each pixel in the image, and improves the visibility and detail contrast of the image.

[0008] The specific method of performing data recovery on the video image in step 2 is: The dark channel of the image is calculated, the atmospheric light and transmittance are estimated, and the transmittance is refined using a guided filter to restore the true color and clarity of the image.

[0009] The specific method of using multi-threaded pipeline processing in step 3 is: Image enhancement and image restoration are processed in different threads respectively. After the processing is completed, the images are timestamped to ensure the synchronization of image processing.

[0010] The specific method of using the weighted image addition algorithm in step 4 is: The weighted image addition algorithm is used to set the weights as follows: the weight of the histogram equalization image is 0.8, and the weight of the dark channel prior method image is 0.2.

[0011] The specific method of outputting the video in step 4 is: The processed image frames are linked into a video stream through the video encoding library and output in the form of a video stream or file.

[0012] Compared with the prior art, the present invention has the following beneficial effects: the present invention adopts histogram equalization to improve the brightness and contrast of underground video images in coal mines, effectively enhances the detail expression of underground video images in coal mines, adopts dark channel prior method (DCP) to clearly restore images blurred by dust, restores the true color and clarity, adopts multi-threaded pipeline processing mode, realizes efficient connection of processing steps, and ensures the stability of real-time processing; the present invention fuses different processing results through weighted image addition operation, realizes the best combination of image enhancement and restoration, and transmits relevant data to the display in real time through wireless network. The method comprehensively improves the working efficiency of underground roadheader drivers and provides a more efficient and reliable solution for safe operation in coal mines. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The present invention will be further described below in conjunction with the accompanying drawings: Figure 1 It is a flow chart of the steps of the video image dust filtering monitoring method of the present invention; Figure 2 It is a schematic diagram of the flow of buffer data when the present invention processes video images; Figure 3 is a flow chart of calculating parameters for image processing in an embodiment of the present invention; Figure 4 3 is a diagram showing the effect of dust filtering of a video image in an embodiment of the present invention. DETAILED DESCRIPTION

[0014] like Figures 1 to 4 As shown, the present invention is used for auxiliary monitoring of the operation of the tunnel boring machine. In order to realize the safety monitoring and automatic control of coal mine production, a coal mine underground video dust filtration monitoring method based on model fusion is provided. The method adopts weighted image addition to fuse image enhancement and image restoration, and uses a wireless base station to transmit the video image to the main driver. It mainly collects the on-site image based on a high-definition camera, and completes the image processing by using the coal mine underground video dust filtration method based on model fusion. Wireless communication is used to assist the main driver in tunneling operations, which can save the assistant driver's position. The method can not only maintain the clarity and contrast of the video image in a complex environment, but also effectively improve the real-time and accuracy of the video monitoring system in a complex environment, realize the visualization of the tunneling working face, and help to improve the safety and efficiency of underground operations.

[0015] In an embodiment of the present invention, in order to achieve the corresponding monitoring effect, four dust removal cameras, two wireless base stations and display devices are added to the working area of ​​the tunnel boring machine, combined with video dust filtering technology, wireless communication technology, etc., through image enhancement and image restoration means, it is possible to improve the brightness and contrast of the collected image through histogram equalization under extremely low visibility conditions, and then use the dark channel prior method to remove the influence of dust, and adopt multi-threaded parallel processing to fuse the results of image enhancement and image restoration through weighted image addition to generate a high-quality dust-filtered image. Finally, two wireless base stations are set up to complete efficient and real-time wireless communication. The advantage is that the images collected by the camera can be processed by video dust filtering, which improves the clarity and smoothness of the video. Finally, the wireless network is used to transmit the images to the main driver in real time, ensuring the real-time nature of image transmission. It can replace the original function of the co-driver to assist in the operation, and can greatly improve the operation efficiency and reduce safety risks.

[0016] Furthermore, the present invention proposes a coal mine underground video dust filtration monitoring method based on model fusion. The method realizes efficient dust filtration of coal mine underground videos by introducing histogram equalization, dark channel prior method and weighted image addition operation, aiming to solve the problems of low visibility and unclear video of key parts of tunnel boring machines in the complex environment of coal mine underground. To implement this solution, camera video monitoring and AI intelligent analysis auxiliary system are adopted. The video monitoring mainly adopts four self-cleaning cameras and displays, which are arranged on the front and middle sides of the tunnel boring machine respectively to observe the position of the shovel and cutting head. The AI ​​position analysis auxiliary system mainly includes computers and analysis terminals for image processing, information transmission base stations and background analysis.

[0017] The solution for processing the collected video images of the present invention comprises the following steps: Step 1: Image enhancement: Use histogram equalization technology to redistribute the grayscale distribution of the downhole image; this method calculates the grayscale histogram of the image and remaps the grayscale distribution of the image to the full range through the cumulative distribution function (CDF) to balance the brightness and contrast of the image; specifically, the new grayscale value of each pixel is calculated and assigned to the target histogram to improve the visibility and detail contrast of the image; Step 2: Image restoration: Use the dark channel prior (DCP) method to restore the image. This method is based on the atmospheric scattering model. By calculating the minimum value of the dark channel in the image, the transmittance and atmospheric light of the image are inferred, the blurring effect caused by dust is removed, and the real information of the image is restored, making the key areas underground more clearly visible. Step 3: Multi-threaded pipeline processing: Through multi-threaded pipeline processing, the two steps of image enhancement and image restoration are processed in parallel, so that two dust filtering processes can be performed on the video image at the same time, and the processing results of the same sub-frame image are matched, laying the foundation for realizing multi-result fusion; Step 4: Fusion of multiple results: Use weighted image addition operation to fuse the images processed by histogram equalization and dark channel prior method. This method assigns appropriate weights to different image processing results, and then weightedly adds them pixel by pixel, thereby combining the advantages of both to generate a final image with high brightness, contrast and clarity, ensuring the balance and consistency of dust filtering effect. Step 5: Wireless communication: Since the tunnel boring machine at the underground excavation working face is in a state of frequent movement during operation, a wireless network is required for signal transmission. For this purpose, two base stations need to be set up, one installed on the tunnel boring machine body and the other installed on the tunnel. The two base stations transmit data to each other to achieve wireless signal transmission.

[0018] The present invention performs efficient dust filtering on underground video images through multi-model fusion and multi-threaded concurrent processing, and transmits video image information of key viewing angles to the control center through wireless communication technology; first, histogram equalization is used to improve the brightness and contrast of images collected by high-definition cameras; secondly, image restoration is performed through the dark channel prior method (DCP) to remove the blurring effect caused by dust; image enhancement and restoration processing steps are performed in parallel, and the multi-threaded pipeline ensures high efficiency and real-time performance; the processing results are fused through weighted image addition, and high-quality images are generated by combining the advantages of both; finally, the images are transmitted to the main driver through a wireless communication base station. This method improves the clarity of underground video images, ensures the consistency of dust filtering effects, replaces the monitoring function of the assistant driver, and is suitable for underground excavation operations in coal mines.

[0019] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for monitoring underground coal mine dust filtration using video based on model fusion, characterized in that: The monitoring steps include the following: Step 1: The video data of the underground working area environment is collected in real time through the camera installed in the coal mine, and the video data is uploaded to the server set up in the tunnel in real time. The server first captures the image frame from the underground video of the coal mine, and timestamps the image frame, and then pre-processes each image frame with histogram equalization to improve the brightness and contrast of the collected video image to achieve image enhancement; Step 2: The server further uses a dark channel prior method to recover the video image data, remove the blurring effect caused by dust on the image, and restore the true clarity of the image; Step 3: The server processes the image enhancement and image restoration steps in parallel by adopting a multi-threaded pipeline processing method; Step 4: The server uses a weighted image addition algorithm to fuse the results of image enhancement and image restoration to combine the advantages of both, generate multiple final images with optimized brightness, contrast, and clarity, and connect the final images into a video for output; Step 5: The server sends the output video to the base station controller on the tunnel boring machine through the wireless network, and the tunnel boring machine driver views it in real time.

2. According to the model fusion-based video dust filtration monitoring method for underground coal mines according to claim 1, it is characterized by: When performing the image enhancement operation, the step 1 specifically calculates the grayscale histogram and cumulative distribution function of the image, maps the grayscale value of each pixel in the image, and improves the visibility and detail contrast of the image.

3. The method for monitoring underground coal mine dust filtration based on model fusion according to claim 1, characterized in that: The specific method of performing data recovery on the video image in step 2 is: The dark channel of the image is calculated, the atmospheric light and transmittance are estimated, and the transmittance is refined using a guided filter to restore the true color and clarity of the image.

4. The method for monitoring underground coal mine dust filtration based on model fusion according to claim 1, characterized in that: The specific method of using multi-threaded pipeline processing in step 3 is: Image enhancement and image restoration are processed in different threads respectively. After the processing is completed, the images are timestamped to ensure the synchronization of image processing.

5. The method for monitoring underground coal mine dust filtration based on model fusion according to claim 1, characterized in that: The specific method of using the weighted image addition algorithm in step 4 is: The weighted image addition algorithm is used to set the weights as follows: the weight of the histogram equalization image is 0.8, and the weight of the dark channel prior method image is 0.

2.

6. The method for monitoring underground coal mine dust filtration based on model fusion according to claim 1, characterized in that: The specific method of outputting the video in step 4 is: The processed image frames are linked into a video stream through the video encoding library and output in the form of a video stream or file.

Images