Submerged chain conveyor slag quantity determination method and device based on image recognition and submerged chain conveyor

Abstract

The invention provides a slag conveyor slag quantity determining method and device based on image recognition and a slag conveyor, and belongs to the technical field of image recognizing.The method comprises the steps that a climbing video image of the slag conveyor is collected through a camera; judging whether the frame queue of the ramp climbing video image is empty or not, if not, acquiring a video frame of the ramp climbing video image, and marking a timestamp to obtain a to-be-identified image; whether the number of the obtained video frames is larger than zero or not is determined, and if not, a scraper in the to-be-recognized image is recognized; identifying a dynamic region of interest in the blade; dividing the dynamic region of interest into a preset number of grid regions, and determining the slag content in each grid region; compared with a laser scanner, the method has the advantages that the cost is effectively reduced due to the fact that images are directly collected through the camera, image recognition is carried out in a video frame number frame mode, the dynamic interested area is combined, the calculation amount is greatly reduced, and the slag amount calculation accuracy is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of image recognition, and in particular to a method and device for determining the slag amount of a slag remover based on image recognition, and a slag remover. Background Art

[0002] In order to improve economic efficiency and reduce wear during the normal operation of the boiler slag remover, the rotation speed needs to be increased or decreased according to the amount of slag. At present, the slag amount on the scraper on the ramp of the slag remover is mostly scanned by a laser scanner to obtain multi-frame sub-point cloud data of the slag, and then the multi-frame sub-point cloud data is processed to calculate the volume of the slag.

[0003] However, laser scanners are very expensive compared to high-definition cameras. The entire scanning process of laser scanners takes a certain amount of time, so there is a delay in the generation of multi-frame sub-point cloud data sets. In addition, due to the large amount of data sets, the computer computing power is directly large, so the hardware requirements for the computer are also higher. Summary of the invention

[0004] The present invention provides a method and device for determining the slag amount of a slag scoop machine based on image recognition, and a slag scoop machine, so as to solve the defects of the slag amount recognition of the slag scoop machine in the prior art, which has large calculation amount and high cost.

[0005] In a first aspect, the present invention provides a method for determining the amount of slag in a slag remover based on image recognition, comprising:

[0006] The video image of the slag picker climbing the ramp is collected by the camera;

[0007] Determine whether the frame queue of the climbing road video image is empty, and if not, obtain the video frame of the climbing road video image, mark the timestamp, and obtain the image to be identified;

[0008] Determine whether the number of the acquired video frames is greater than zero, and if not, identify the scraper in the image to be identified;

[0009] identifying a dynamic region of interest in the scraper;

[0010] Dividing the dynamic region of interest into a preset number of grid areas, and determining the slag content in each of the grid areas;

[0011] The slag content in each of the grid areas is counted to determine the slag volume of the slag remover.

[0012] According to a method for determining the amount of slag in a slag remover based on image recognition provided by the present invention, the step of identifying a dynamic region of interest in the scraper includes:

[0013] Obtain the scraper coordinates through Hough change straight line detection;

[0014] Through the scraper coordinates, the slag volume is obtained to identify the selected dynamic area of ​​interest.

[0015] According to a method for determining the slag amount of a slag remover based on image recognition provided by the present invention, the scraper coordinates are obtained by Hough change straight line detection, comprising:

[0016] After obtaining the initial scraper coordinates through Hough change straight line detection, start counting frames;

[0017] When the result of the number of frames is greater than a preset value, the scraper coordinates are reacquired.

[0018] According to a method for determining the amount of slag in a slag remover based on image recognition provided by the present invention, the step of identifying a scraper in the image to be identified includes:

[0019] Performing preprocessing of perspective change, filtering, graying and adaptive equalization on the image to be identified to obtain a standard image;

[0020] The standard image is subjected to edge detection, anti-glare and area stitching to determine the scraper in the standard image.

[0021] According to a method for determining the slag amount of a slag remover based on image recognition provided by the present invention, the determining the slag content in each grid area includes:

[0022] Determine the pure white pixel set and the pure black pixel set in each grid area;

[0023] The proportion of pure black pixels in each of the grid areas is determined. When the proportion is greater than a preset ratio, the grid area is determined to contain slag, otherwise it is determined to contain no slag.

[0024] A method for determining the amount of slag in a slag remover based on image recognition according to the present invention also includes:

[0025] If the number of video frames is greater than zero, the number of video frames is increased by 1 until the number of video frames is greater than 15 frames, and then the number of video frames is restored to 0.

[0026] According to a method for determining the amount of slag in a slag scoop machine based on image recognition provided by the present invention, after determining the amount of slag in the slag scoop machine, the method further comprises:

[0027] storing the slag amount in a database queue;

[0028] Sending the database queue to a DCS control system via a network switch and a PLC;

[0029] The DCS control system sends a slag scoop machine control instruction to the slag scoop machine to adjust the rotation speed of the slag scoop machine. The slag scoop machine control instruction is determined based on the slag amount in the database queue.

[0030] A method for determining the amount of slag in a slag remover based on image recognition according to the present invention also includes:

[0031] When at least one of the ramp video image information, the network switch information, the PLC information and the DCS control system information cannot be collected, an alarm is issued and a log file is recorded;

[0032] The cause of the fault is determined through the alarm prompt and the log file.

[0033] In a second aspect, the present invention further provides a device for determining the amount of slag in a slag removal machine based on image recognition, comprising:

[0034] A collection module, used to collect video images of the slag scoop machine climbing the ramp through a camera;

[0035] A judging module, used for judging whether the frame queue of the climbing road video image is empty, and if not, obtaining the video frame of the climbing road video image and marking the timestamp to obtain the image to be identified;

[0036] A determination module, used to determine whether the number of the acquired video frames is greater than zero, and if not, identify the scraper in the image to be identified;

[0037] An identification module, for identifying a dynamic region of interest in the scraper;

[0038] A division module, used for dividing the dynamic region of interest into a preset number of grid areas, and determining the slag content in each of the grid areas;

[0039] The statistical module is used to count the slag content in each of the grid areas and determine the slag amount of the slag remover.

[0040] In a third aspect, the present invention further provides a slag scavenger, which is used to execute the method for determining the slag amount of the slag scavenger based on image recognition as described in any one of the above.

[0041] In a fourth aspect, the present invention further provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method for determining the slag amount of a slag removal machine based on image recognition as described in any one of the above is implemented.

[0042] In a fifth aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the above-described methods for determining the amount of slag in a slag removal machine based on image recognition.

[0043] In a sixth aspect, the present invention further provides a computer program product, comprising a computer program, which, when executed by a processor, implements any of the above-mentioned methods for determining the slag amount of a slag removal machine based on image recognition.

[0044] The present invention provides a method, device and slag removal machine for determining the slag amount of a slag removal machine based on image recognition. The method comprises: collecting a climbing video image of the slag removal machine through a camera; judging whether a frame queue of the climbing video image is empty, if not, obtaining a video frame of the climbing video image, and marking a timestamp to obtain an image to be identified; determining whether the number of acquired video frames is greater than zero, if not, identifying a scraper in the image to be identified; identifying a dynamic region of interest in the scraper; dividing the dynamic region of interest into a preset number of grid areas, and determining the slag content in each grid area; counting the slag content in each grid area to determine the slag amount of the slag removal machine. Compared with a laser scanner, directly collecting images through a camera effectively reduces costs, and image recognition is performed by counting video frames, combined with the dynamic region of interest, which greatly reduces the amount of calculation and improves the accuracy of slag amount calculation. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0046] Figure 1 is a flow chart of a method for determining the amount of slag in a slag remover based on image recognition provided in this embodiment;

[0047] Figure 2 yes Figure 1 Schematic diagram of corresponding hardware devices and connection relationships;

[0048] Figure 3 is a structural schematic diagram of a slag amount determination device for a slag remover based on image recognition provided in this embodiment;

[0049] Figure 4 It is a schematic diagram of the structure of the electronic device provided in this embodiment. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0051] Figure 1 is a flow chart of a method for determining the amount of slag from a slag remover based on image recognition provided in this embodiment, Figure 2 yes Figure 1 Schematic diagram of the corresponding hardware devices and connection relationships.

[0052] like Figure 1 As shown, the method for determining the slag amount of a slag dredger based on image recognition provided by an embodiment of the present invention can be performed by a slag amount recognition computing platform, and the method mainly includes the following steps:

[0053] 101. Collect video images of the slag picker climbing the ramp through the camera.

[0054] In a specific implementation process, first follow Figure 2 As shown in the figure, the hardware system for image recognition processing and related control is built, including a high-definition camera, a network switch, an image recognition platform, a programmable logic controller (PLC) and a distributed control system (DCS). Among them, the high-definition camera is facing the slag scoop machine, and the collected video is communicated to the image recognition platform through the network switch. The image recognition platform is connected to the PLC through the network switch, and the DCS collects and controls the I / O points of the PLC in a hard-wired manner, reducing the network security risk of the DCS control system.

[0055] The process of the camera collecting the climbing video image of the slag removal machine can be automatically run after the system is started, and the climbing video image can be collected in real time. The climbing video image refers to the image of the slag removal machine during transportation. The climbing image can more accurately reflect the slag removal amount of the slag removal machine.

[0056] Collecting video images of the ramp includes determining the specific area that the camera needs to cover, including the starting point, midpoint, end point, and possible bottlenecks or high-risk areas of the ramp. If complete coverage is not possible, multiple cameras are used to collect different areas separately, and the images collected in different areas are fused to obtain the target area image.

[0057] At the same time, based on the on-site environment, such as light conditions, distance and viewing angle requirements, choose the appropriate camera type (such as fixed focal length, variable focal length, infrared night vision, etc.). Choose a high enough resolution and frame rate to capture clear images and smooth videos. Considering that the working environment of the slag loader may be relatively harsh, choose a camera with a high waterproof and dustproof grade.

[0058] After the camera is selected and installed, the camera performance needs to be tested: Test the camera's imaging effect under different light and weather conditions to ensure that the video image is clear and smooth. Adjust the camera's angle, focal length and other parameters according to the actual monitoring effect to obtain the best monitoring effect. Adjust the network settings to optimize the transmission speed and stability of the video data. Ensure that the ramp video image can be clearly captured.

[0059] 102. Determine whether the frame queue of the ramp video image is empty. If it is not empty, obtain the video frame of the ramp video image and mark the time stamp to obtain the image to be identified.

[0060] If the frame queue is empty, it indicates that no useful information has been collected. After a delay of 1s, it is re-judged whether the frame queue is empty. When the frame queue is not empty, it indicates that a valid climbing video image has been collected. At this time, obtaining the video frame of the climbing video image can be understood as collecting a certain frame of the picture for image processing, and marking the timestamp at the same time to distinguish different video frames, thereby obtaining images to be identified one by one.

[0061] The image to be identified refers to an image that needs to be analyzed and processed to determine whether it contains slag.

[0062] 103. Determine whether the number of acquired video frames is greater than zero. If not, identify the scraper in the image to be identified.

[0063] After obtaining the image to be identified, the scraper in the image to be identified is identified. Before identification, a condition confirmation is required, that is, whether the number of video frames is greater than zero. Only when it is not greater than zero, the scraper is identified. If the number of video frames is greater than zero, the count is increased by 1 until the number of video frames is greater than 15 frames, and then the number of video frames is restored to 0. That is, it is determined whether the count is greater than 0. If it is greater than 0, the count is increased by 1 until the count is greater than 15, and the count is set to 0; if the count is not greater than 0, the scraper is identified.

[0064] The process of identifying the scraper is to pre-process the image to be identified by perspective change, filtering, graying and adaptive equalization to obtain a standard image; perform edge detection, anti-backlighting and area stitching on the standard image to determine the scraper in the standard image.

[0065] 104. Identify dynamic regions of interest in a scraper.

[0066] The region of interest (ROI) selected for slag volume identification is dynamic, that is, the ROI is variable and not unique. Specifically, the scraper coordinates are obtained through Hough change straight line detection; through the scraper coordinates, the dynamic region of interest selected for slag volume identification is obtained. Among them, the scraper coordinates are obtained through Hough change straight line detection, including: after the initial scraper coordinates are obtained through Hough change straight line detection, start counting frames; when the frame counting result is greater than the preset value of 15 frames, the scraper coordinates are re-obtained.

[0067] Since ROI changes with the scraper coordinates, when the scraper coordinates change, ROI changes accordingly, thus ensuring the accuracy of the ROI area and avoiding the drawbacks of large calculation errors caused by fixed ROI and scraper interference. At the same time, since new scraper coordinates will be obtained only after the number of frames exceeds 15, it ensures that the amount of computer calculation is greatly reduced and the computer memory space is saved.

[0068] 105. Divide the dynamic region of interest into a preset number of grid areas, and determine the slag content in each grid area.

[0069] The region of interest is divided into 100 grid areas, and the pure white pixel set and the pure black pixel set in each grid area are determined. Pure white indicates no slag, and pure black indicates slag. The proportion of pure black pixels in each grid area is determined. When the proportion is greater than the preset ratio, the grid area is determined to contain slag, otherwise it is determined to contain no slag.

[0070] Specifically, it is determined how many grids the dynamic region of interest is to be divided into. The preset number can be determined based on actual needs, image resolution, processing power, etc. For example, a large ROI can be divided into 4x4, 8x8 or more grids.

[0071] Determine the ROI boundary: First, the ROI boundary needs to be clearly defined, which can be achieved through image processing technology (such as edge detection, image segmentation, etc.) or determined through manual annotation.

[0072] Apply partitioning strategy: According to the selected partitioning strategy, including uniform partitioning or non-uniform partitioning, grid partitioning is performed in the ROI area. If it is uniform partitioning, the grid generation algorithm can be used; if it is non-uniform partitioning, the partitioning area needs to be determined according to the position coordinates of the ROI area. If the position coordinates represent the two sides of the scraper, the grid area is larger, and if the position coordinates represent the middle of the scraper, the grid area is smaller. The grid area can be linearly adjusted and optimized according to the distance from the middle of the scraper to ensure the rationality of the grid area.

[0073] When dividing the grids, it should be ensured that each grid is relatively independent to avoid mutual interference or overlap. The number of divided grids should be moderate, which can not only meet the needs of analysis or processing, but also ensure the efficiency of data processing. At the same time, since the ROI area may change dynamically (such as moving targets in surveillance videos), the division strategy should have a certain degree of flexibility and adaptability so that the grid division can be adjusted in time to adapt to new situations.

[0074] 106. Count the slag content in each grid area and determine the slag amount of the slag removal machine.

[0075] Since the slag content of each grid area can be determined according to the proportion of black, after determining the proportion of each grid area, the slag content of each grid area in all grid areas is statistically calculated to obtain the overall slag content.

[0076] The method for determining the slag amount of the slag scoop machine of this embodiment reduces the labor intensity of the operating inspectors, effectively avoids the occurrence of overload tripping or equipment damage due to the large amount of slag caused by the operating personnel due to untimely monitoring, and ensures the safety of the equipment. The wear of the scraper of the slag scoop machine is effectively reduced, and the service life of the slag scoop machine is increased. By adjusting the speed of the slag scoop machine, the full output operation of the slag scoop machine with no load or small load is avoided, which saves electricity consumption and improves economy. PLC, as an image recognition platform, is hardware isolated from the DCS control system, which effectively reduces the network security risk of the DCS control system.

[0077] The image recognition algorithm uses a frame-by-frame algorithm, which greatly reduces the amount of computer calculations and saves computer memory space. The ROI selected by the program's slag volume recognition is dynamic. The scraper coordinates within the ROI selected by the scraper recognition are obtained through Hough change straight line detection, and then the ROI selected by the slag volume recognition is obtained through the scraper coordinates. The drawbacks of large calculation errors caused by fixed ROI and scraper interference are avoided. The regional stitching algorithm completes and stitches the detected scattered areas, which can effectively reduce the drawbacks of poor adaptability of the program threshold that has not been through the regional stitching algorithm due to light intensity. The image recognition algorithm is simple and practical, with low computer calculations and low requirements for computer hardware. In addition, high-definition cameras also greatly reduce costs compared to laser scanners. Using PLC as the image recognition platform and hardware isolation from the DCS control system effectively reduces the network security risk of the DCS control system.

[0078] Furthermore, based on the above-mentioned embodiments, after determining the slag amount of the slag recovery machine in this embodiment, it also includes: storing the slag amount in a database queue; sending the database queue to a DCS control system through a network switch and a PLC; the DCS control system sends a slag recovery machine control instruction to the slag recovery machine to adjust the rotation speed of the slag recovery machine, and the slag recovery machine control instruction is determined based on the slag amount in the database queue.

[0079] Specifically, in the whole slag volume determination process, multi-threading technology is adopted, and the main thread is responsible for starting, monitoring and controlling all sub-threads, including video stream acquisition sub-thread, communication sub-thread, calculation sub-thread, recording sub-thread, memory mapping sub-thread, etc. The program response speed is improved, and the time for each frame image recognition and processing is shortened.

[0080] After determining the amount of slag, the data is transmitted to the DCS control system through the network switch and PLC for reference by the staff. The staff can then manually adjust the speed and other working parameters of the slag remover according to the data or the DCS can automatically adjust the speed of the slag remover through logical configuration.

[0081] When at least one of the ramp video image information, network switch information, PLC information and DCS control system information cannot be collected, an alarm prompt is issued and a log file is recorded; the cause of the fault is determined through the alarm prompt and log file, which facilitates maintenance and inspection by staff.

[0082] Based on the same general inventive concept, the present invention also protects a device for determining the slag amount of a slag recovery machine based on image recognition. The device for determining the slag amount of a slag recovery machine based on image recognition provided by the present invention is described below. The device for determining the slag amount of a slag recovery machine based on image recognition described below and the method for determining the slag amount of a slag recovery machine based on image recognition described above can refer to each other.

[0083] Figure 3 It is a structural schematic diagram of the slag amount determination device of the slag remover based on image recognition provided in this embodiment.

[0084] like Figure 3 As shown, this embodiment provides a slag amount determination device for a slag remover based on image recognition, comprising:

[0085] The acquisition module 301 is used to acquire the video image of the slag scoop machine climbing the ramp through the camera;

[0086] The judging module 302 is used to judge whether the frame queue of the video image of the climbing road is empty. If it is not empty, the video frame of the video image of the climbing road is obtained and the time stamp is marked to obtain the image to be identified;

[0087] The determination module 303 is used to determine whether the number of acquired video frames is greater than zero, and if not, identify the scraper in the image to be identified;

[0088] An identification module 304 for identifying a dynamic region of interest in the scraper;

[0089] A division module 305 is used to divide the dynamic region of interest into a preset number of grid areas and determine the slag content in each grid area;

[0090] The statistical module 306 is used to count the slag content in each grid area and determine the slag amount of the slag remover.

[0091] Furthermore, the identification module 304 in this embodiment is specifically used for:

[0092] Obtain the scraper coordinates through Hough change straight line detection;

[0093] Through the scraper coordinates, the slag volume is obtained to identify the selected dynamic area of ​​interest.

[0094] Furthermore, the identification module 304 in this embodiment is also specifically used for:

[0095] After obtaining the initial scraper coordinates through Hough change straight line detection, start counting frames;

[0096] When the result of the number of frames is greater than a preset value, the scraper coordinates are reacquired.

[0097] Furthermore, the determination module 303 in this embodiment is specifically used to:

[0098] Performing preprocessing of perspective change, filtering, graying and adaptive equalization on the image to be identified to obtain a standard image;

[0099] The standard image is subjected to edge detection, anti-glare and area stitching to determine the scraper in the standard image.

[0100] Furthermore, the statistics module 306 in this embodiment is specifically used for:

[0101] Determine the pure white pixel set and the pure black pixel set in each grid area;

[0102] The proportion of pure black pixels in each of the grid areas is determined. When the proportion is greater than a preset ratio, the grid area is determined to contain slag, otherwise it is determined to contain no slag.

[0103] Furthermore, the determination module 303 in this embodiment is further configured to:

[0104] If the number of video frames is greater than zero, the number of video frames is increased by 1 until the number of video frames is greater than 15 frames, and then the number of video frames is restored to 0.

[0105] Furthermore, the present embodiment also includes: a control module, which is used to:

[0106] storing the slag amount in a database queue;

[0107] Sending the database queue to a DCS control system via a network switch and a PLC;

[0108] The DCS control system sends a slag scoop control instruction to the slag scoop to adjust the rotation speed of the slag scoop, wherein the slag scoop control instruction is determined based on the slag amount in the database queue;

[0109] When at least one of the ramp video image information, the network switch information, the PLC information and the DCS control system information cannot be collected, an alarm is issued and a log file is recorded;

[0110] The cause of the fault is determined through the alarm prompt and the log file.

[0111] The present invention also protects a slag scavenger, which is used to execute the slag amount determination method of the slag scavenger based on image recognition as described in any of the above embodiments.

[0112] Figure 4 It is a schematic diagram of the structure of the electronic device provided in this embodiment.

[0113] like Figure 4 As shown, the electronic device may include: a processor 410, a communication interface 420, a memory 430 and a communication bus 440, wherein the processor 410, the communication interface 420 and the memory 430 communicate with each other through the communication bus 440. The processor 410 may call the logic instructions in the memory 430 to execute the method for determining the slag amount of the slag scavenger based on image recognition, the method comprising: collecting the ramp video image of the slag scavenger through a camera; judging whether the frame queue of the ramp video image is empty, if not empty, obtaining the video frame of the ramp video image, marking the timestamp, and obtaining the image to be identified; determining whether the number of the obtained video frames is greater than zero, if not greater than zero, identifying the scraper in the image to be identified; identifying the dynamic interest region in the scraper; dividing the dynamic interest region into a preset number of grid areas, determining the slag content in each grid area; counting the slag content in each grid area, and determining the slag amount of the slag scavenger.

[0114] In addition, the logic instructions in the above-mentioned memory 430 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0115] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the above-mentioned methods to collect the climbing video image of the slag recovery machine through a camera; determine whether the frame queue of the climbing video image is empty, if not empty, obtain the video frame of the climbing video image, and mark the timestamp to obtain the image to be identified; determine whether the number of acquired video frames is greater than zero, if not greater than zero, identify the scraper in the image to be identified; identify the dynamic area of ​​interest in the scraper; divide the dynamic area of ​​interest into a preset number of grid areas, and determine the slag content in each grid area; count the slag content in each grid area, and determine the slag amount of the slag recovery machine

[0116] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the above-mentioned methods to collect a video image of a slag recovery machine climbing a ramp through a camera; determine whether the frame queue of the climbing ramp video image is empty, and if not, obtain a video frame of the climbing ramp video image and mark it with a timestamp to obtain an image to be identified; determine whether the number of acquired video frames is greater than zero, and if not, identify a scraper in the image to be identified; identify a dynamic region of interest in the scraper; divide the dynamic region of interest into a preset number of grid areas, and determine the slag content in each grid area; count the slag content in each grid area, and determine the slag amount of the slag recovery machine

[0117] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0118] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0119] 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 make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for determining the amount of slag in a slag remover based on image recognition, characterized in that: include: The video image of the slag picker climbing the ramp is collected by the camera; Determine whether the frame queue of the climbing road video image is empty, and if not, obtain the video frame of the climbing road video image, mark the timestamp, and obtain the image to be identified; Determine whether the number of the acquired video frames is greater than zero, and if not, identify the scraper in the image to be identified; identifying a dynamic region of interest in the scraper; Dividing the dynamic region of interest into a preset number of grid areas, and determining the slag content in each of the grid areas; The slag content in each of the grid areas is counted to determine the slag volume of the slag remover.

2. The method for determining the slag amount of a slag remover based on image recognition according to claim 1, characterized in that: The identifying of a dynamic region of interest in the scraper comprises: Obtain the scraper coordinates through Hough change straight line detection; Through the scraper coordinates, the slag volume is obtained to identify the selected dynamic area of ​​interest.

3. The method for determining the slag amount of a slag remover based on image recognition according to claim 2, characterized in that: The method of obtaining the scraper coordinates by Hough change straight line detection includes: After obtaining the initial scraper coordinates through Hough change straight line detection, start counting frames; When the result of the number of frames is greater than a preset value, the scraper coordinates are reacquired.

4. The method for determining the slag amount of a slag remover based on image recognition according to claim 1, characterized in that: The step of identifying a scraper in the image to be identified includes: Performing preprocessing of perspective change, filtering, graying and adaptive equalization on the image to be identified to obtain a standard image; The standard image is subjected to edge detection, anti-glare and area stitching to determine the scraper in the standard image.

5. The method for determining the slag amount of a slag remover based on image recognition according to claim 1, characterized in that: The determining of the slag content in each of the grid areas comprises: Determine the pure white pixel set and the pure black pixel set in each grid area; The proportion of pure black pixels in each of the grid areas is determined. When the proportion is greater than a preset ratio, the grid area is determined to contain slag, otherwise it is determined to contain no slag.

6. The method for determining the slag amount of a slag remover based on image recognition according to any one of claims 1 to 5, characterized in that: Also includes: If the number of video frames is greater than zero, the number of video frames is increased by 1 until the number of video frames is greater than 15 frames, and then the number of video frames is restored to 0.

7. The method for determining the slag amount of a slag remover based on image recognition according to any one of claims 1 to 5, characterized in that: After determining the slag amount of the slag scooping machine, the method further includes: Storing the slag amount in a database queue; Sending the database queue to a DCS control system via a network switch and a PLC; The DCS control system sends a slag scoop machine control instruction to the slag scoop machine to adjust the rotation speed of the slag scoop machine. The slag scoop machine control instruction is determined based on the slag amount in the database queue.

8. The method for determining the slag amount of a slag remover based on image recognition according to claim 7, characterized in that: Also includes: When at least one of the ramp video image information, the network switch information, the PLC information and the DCS control system information cannot be collected, an alarm is issued and a log file is recorded; The cause of the fault is determined through the alarm prompt and the log file.

9. A device for determining the amount of slag in a slag dredger based on image recognition, characterized in that: include: A collection module, used to collect video images of the slag scoop machine climbing the ramp through a camera; A judging module, used for judging whether the frame queue of the climbing road video image is empty, and if not, obtaining the video frame of the climbing road video image and marking the timestamp to obtain the image to be identified; A determination module, used to determine whether the number of the acquired video frames is greater than zero, and if not, identify the scraper in the image to be identified; An identification module, for identifying a dynamic region of interest in the scraper; A division module, used for dividing the dynamic region of interest into a preset number of grid areas, and determining the slag content in each of the grid areas; The statistical module is used to count the slag content in each of the grid areas and determine the slag amount of the slag remover.

10. A slag scooping machine, characterized in that: The slag scoop machine is used to execute the slag amount determination method of the slag scoop machine based on image recognition as described in any one of claims 1-8.

Images