Visual detection method and device for chain breakage fault of flight conveyor

Abstract

The present application relates to the technical field of scraper conveyor fault detection, and specifically provides a visual detection method and device for chain breakage fault of a scraper conveyor, aiming to solve the problem of existing chain breakage fault detection methods that sensors are prone to wear, have low sensitivity and poor ability to resist harsh environments. To this end, the visual detection method provided by the present application comprises the following steps: S10: acquiring an image of the scraper conveyor; S20: determining whether the image is empty, if yes, stopping the program running, and if no, executing S30; S30: performing first processing on the image to obtain a first image; S40: determining whether the first image is the first frame, if yes, executing S30, and if no, executing S50; S50: performing second processing on the first image to obtain a second image and acquiring the number n of moving objects; S60: determining whether n is zero, if yes, executing S70, and if no, executing S30; S70: acquiring a chain wheel rotating speed signal of the scraper conveyor; S80: determining whether the chain wheel rotating speed is zero, if yes, determining to stop the machine, and if no, alarming a chain breakage fault.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of scraper conveyor fault detection, and particularly provides a visual detection method and device for chain breakage fault of a scraper conveyor. BACKGROUND

[0002] The existing scraper conveyor is a conveyor that uses a scraper chain to pull and transport bulk materials in a trough. In the current coal mining face, the scraper conveyor not only serves to transport coal and materials, but also serves as the running track of the coal mining machine, so it becomes an indispensable main equipment in the modern coal mining process. The scraper conveyor can maintain continuous operation, and the production can proceed normally. Otherwise, the entire coal mining face will be in a state of production stoppage, causing the entire production to be interrupted. The common fault of the scraper conveyor is chain breakage. Therefore, timely detection of the chain breakage fault is an important measure to ensure production tasks.

[0003] The common detection methods for chain breakage fault of the scraper conveyor can be roughly divided into two categories according to whether contact is involved: contact type and non-contact type. The common contact type detection methods include chain tension detection method, tongue plate type detection method and sprocket stress detection method, and the common non-contact type detection methods include sprocket speed detection method and Hall sensor detection method.

[0004] For the contact type detection method, due to wear and tear and collision of coal and gangue, etc., such sensors generally have the defects of easy wear, low sensitivity, poor resistance to harsh environments, and poor reliability. For the non-contact type detection method, the sprocket speed detection method requires real-time measurement of the speeds of two sprockets, and the difference is used for judgment, which has low sensitivity and cannot detect slight chain faults, and lacks applicability to front and rear double drive type scraper conveyors. The Hall sensor detection method requires that the sensor cannot affect the passage of coal, scraper and coal mining machine, otherwise the sensor is easy to be damaged or worn, so the installation is not very convenient, and the structure is relatively complex.

[0005] Correspondingly, there is a need in the art for a new detection method for chain breakage fault of a scraper conveyor to solve the above problems. SUMMARY

[0006] The present application aims to solve the above technical problems, i.e. to solve the problem of existing chain breakage fault detection methods that sensors are easy to wear, have low sensitivity and poor resistance to harsh environments. To this end, the present application provides a visual detection method for chain breakage fault of a scraper conveyor, which comprises the following steps:

[0007] Step S10: acquiring an image of the scraper conveyor;

[0008] Step S20: determining whether the image is empty, if yes, stopping the program running, and if no, executing step S30;

[0009] Step S30: performing first processing on the image to obtain a first image;

[0010] Step S40: judging whether the first image is the first frame, if yes, performing step S30, if no, performing step S50;

[0011] Step S50: performing second processing on the first image to obtain a second image, and obtaining the number n of moving objects on the second image;

[0012] Step S60: judging whether n is zero, if yes, performing step S70, if no, performing step S30;

[0013] Step S70: obtaining the chain wheel rotating speed signal of the scraper conveyor;

[0014] Step S80: judging whether the chain wheel is running, if yes, determining normal shutdown, if no, performing chain breakage fault alarm program.

[0015] In the specific embodiment of the visual detection method, in the step S10, the first processing includes sequentially performing grayscale processing, normalization processing and histogram equalization processing on the image, and the first image after processing has the characteristic of uniform grayscale probability density distribution.

[0016] In the specific embodiment of the visual detection method, the calculation formula of the normalization processing and the histogram equalization processing is as follows:

[0017]

[0018]

[0019] In the above formula, s represents the pixel value of the image after histogram equalization processing, r represents the pixel value of the image after normalization processing, T(r) represents the transformation function, P r (r) represents the probability density function of the pixel value r of the image after normalization processing.

[0020] In the specific embodiment of the visual detection method, in the step S10, the video information of the scraper conveyor is first obtained, and then the video information is processed to obtain the real-time image.

[0021] In the specific embodiment of the visual detection method, in the step S30, the second processing includes the following steps:

[0022] Step S31: performing motion target detection to obtain a difference image or other motion target information;

[0023] Step S32: intersecting the difference part with the mask to obtain a part of the scraper conveyor in the image, as a first intermediate image;

[0024] Step S33: performing a binaryzation process on the first intermediate image to obtain a second intermediate image;

[0025] Step S34: performing a morphological process on the second intermediate image to obtain a second image.

[0026] In the embodiment of the visual detection method, the method of the binaryzation process is as follows: selecting a proper threshold T, and using the following formula to process each pixel point, covering the pixel point with 255 if the pixel value of the pixel point in the gray image is greater than or equal to the threshold, and the binaryzation result of the pixel point is foreground; covering the pixel point with 0 if the pixel value of the pixel point in the gray image is less than the threshold, and the binaryzation result of the pixel point is background, to obtain a binaryzation image:

[0027]

[0028] In the above formula, p(x, y) represents the pixel value at the pixel point (x, y) after the binaryzation process of the gray image, and gray(x, y) represents the pixel value at the pixel point (x, y) in the gray image.

[0029] In the embodiment of the visual detection method, the method of the morphological process is as follows: sequentially performing an erosion operation and a dilation operation on the second intermediate image.

[0030] In the embodiment of the visual detection method, the motion target detection adopts an inter-frame difference method, an optical flow method or a background subtraction method.

[0031] In the embodiment of the visual detection method, in the step S20, the mask can be manually extracted, or can be automatically extracted through image segmentation or target detection.

[0032] In the case of using the above technical solution, the visual detection method provided by the application overcomes the problems of the existing chain breakage fault detection method, such as easy wear of the sensor, low sensitivity and poor ability to resist harsh environments. Specifically, the image of the scraper conveyor is obtained and processed, it is first judged whether the moving object on the processed image is zero, then the chain wheel speed signal of the scraper conveyor is obtained from the variable frequency converter, the vacuum electromagnetic starter or other chain wheel speed monitoring devices, it is judged whether the scraper conveyor is stopped, and then the two judgment results are combined to determine whether the scraper conveyor has a chain breakage fault.

[0033] The method only needs to acquire the dynamic image of the scraper conveyor and the rotating speed information of the chain wheel to determine whether the scraper conveyor is only in the chain breakage fault state, the image processing related equipment can be arranged at a relatively safe position, the stability of the image processing equipment is improved, and therefore the whole system has strong robustness.

[0034] In addition, the method improves the intelligent level of the scraper conveyor, guarantees the continuity of coal transportation in the working face, improves the production efficiency, reduces the labor intensity of workers, and maximally reduces the influence and loss of coal mine enterprises caused by the chain fault of the scraper conveyor, and has important engineering application value.

[0035] The application also provides a visual detection device for the chain breakage fault of the scraper conveyor, the visual detection device comprises a controller, and the controller is configured to execute the visual detection method in any one of the above embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0036] The preferred embodiments of the application are described below with reference to the drawings, in which:

[0037] Figure 1 is the main step flowchart of the visual detection method provided by the application;

[0038] Figure 2 is the step flowchart of the second processing in the visual detection method provided by the application;

[0039] Figure 3 is a schematic diagram of the visual detection device provided by the application;

[0040] Figure 4 is the step flowchart of the algorithm transplantation into the embedded platform provided by the application. DETAILED DESCRIPTION

[0041] It should be noted that, in the description of the application, the ordinal numbers "first", "second", etc. are only used for the purpose of description, and cannot be understood as indicating or implying relative importance.

[0042] As Figure 1 shown, the application provides a visual detection method for the chain breakage fault of a scraper conveyor, the visual detection method comprises the following steps:

[0043] Step S10: acquiring an image of the scraper conveyor;

[0044] Step S20: judging whether the image is empty, if yes, stopping the program running, and if no, executing step S30;

[0045] Step S30: performing first processing on the image to obtain a first image;

[0046] Step S40: judging whether the first image is the first frame, if yes, executing step S30, if no, executing step S50;

[0047] Step S50: secondly processing the first image to obtain a second image, and acquiring the number n of moving objects on the second image;

[0048] Step S60: judging whether n is zero, if yes, executing step S70, if no, executing step S30;

[0049] Step S70: acquiring the chain wheel rotating speed signal of the scraper conveyor;

[0050] Step S80: judging whether the chain wheel is running, if yes, determining normal shutdown, if no, executing the chain breakage fault alarm program.

[0051] The visual detection method provided by the application overcomes the problems of sensor wear, low sensitivity, poor anti-adverse environment capability and the like in the prior art. Specifically, the image of the scraper conveyor is acquired and processed, it is judged whether the moving object on the processed image is zero, the chain wheel rotating speed signal of the scraper conveyor is obtained from the variable frequency converter, the vacuum electromagnetic starter or other chain wheel rotating speed monitoring device, it is judged whether the scraper conveyor is shutdown, and then the two judgment results are comprehensively determined to determine whether the scraper conveyor has the chain breakage fault.

[0052] The method only needs to acquire the dynamic image of the scraper conveyor and the rotating speed information of the chain wheel to determine whether the scraper conveyor is only in the chain breakage fault state, the image processing related equipment can be arranged at a relatively safe position, the stability of the image processing equipment is improved, and thus the whole system has strong robustness.

[0053] In addition, the method improves the intelligent level of the scraper conveyor, guarantees the continuity of the coal transportation of the working face, improves the production efficiency, reduces the labor intensity of workers, maximally reduces the influence and loss of the coal mine enterprises due to the chain breakage fault of the scraper conveyor, and has important engineering application value.

[0054] Further, in the step S10, the first processing includes sequentially performing the grayscale processing, the normalization processing and the histogram equalization processing on the image, and the first image after the processing has the feature of uniform grayscale probability density distribution.

[0055] In the embodiment, the grayscale processing on the image can reduce the size of the storage data space, and at the same time, can also reduce the amount of data to be operated, improve the speed of subsequent image processing, the normalization processing and the histogram equalization processing are both to enhance the outline of the target in the image, so as to improve the definition and brightness of the image, facilitate the features of the target object in the image to be more obvious, and enhance the subsequent detection effect.

[0056] Further, the calculation formula of the normalization processing is as follows:

[0057]

[0058] In the above formula, L-1 represents a gray value, pixel value of an original image belongs to (0, L-1), P r (r) represents a probability density function of the image pixel value after the normalization processing;

[0059] The relationship between the probability density function of the image pixel value r after the normalization processing and the probability density function of the image pixel value s after the histogram equalization processing is as follows:

[0060] P s (s)d s =P r (r)d r

[0061] The transformation function is obtained by simultaneously integrating both sides:

[0062]

[0063] In the above formula, s represents the image pixel value after the histogram equalization processing, r represents the image pixel value after the normalization processing, T(r) represents the transformation function, P r (r) represents the probability density function of the image pixel value r after the normalization processing.

[0064] Further, in the step S10, video information of the scraper conveyor is acquired first, and then the video information is processed to obtain a real-time image.

[0065] Further, as Figure 2 indicated, in the step S30, the "second processing" includes the following steps:

[0066] Step S31: motion target detection is performed to obtain a difference image or other motion target information;

[0067] Step S32: the difference part is intersected with a mask to obtain a part of the scraper conveyor in the image, which is a first intermediate image;

[0068] Step S33: binaryzation processing is performed on the first intermediate image to obtain a second intermediate image;

[0069] Step S34: morphological processing is performed on the second intermediate image to obtain the second image.

[0070] In the embodiment, in order to reduce redundant calculation and interference of redundant background, prevent the motion of coal, gangue, pedestrians and other equipment or objects from causing misjudgment of the scraper conveyor chain breakage fault, improve the algorithm running speed and recognition accuracy, the specific position of the scraper conveyor in the image is set as a mask, the intersection of the image and the mask is obtained, and the part of the scraper conveyor in the difference image is obtained. The binaryzation processing of the image can present obvious black and white effect of the whole image, thereby greatly reducing the data amount of the image, and finally highlighting the contour of the target. The morphological processing can reduce the interference in the image and improve the accuracy of target detection.

[0071] Further, the binaryzation processing method is: selecting a suitable threshold T, processing each pixel point using the following formula, covering the pixel point with 255 if the pixel value of the pixel point in the gray image is greater than or equal to the threshold value, and the binaryzation result of the pixel point is foreground; covering the pixel point with 0 if the pixel value of the pixel point in the gray image is less than the threshold value, and the binaryzation result of the pixel point is background, and obtaining a binaryzation image:

[0072]

[0073] In the formula, p(x, y) represents the pixel value of the pixel point (x, y) in the gray image after binaryzation processing, and gray(x, y) represents the pixel value of the pixel point (x, y) in the gray image.

[0074] Further, the morphological processing method is: sequentially performing erosion operation and dilation operation on the second intermediate image.

[0075] In the embodiment, the erosion operation can remove the interference information in the image caused by light change, and then the dilation operation is performed on the image to eliminate the excessive erosion of the target contour caused by the erosion operation, so as to ensure the extraction accuracy of the target.

[0076] Further, the motion target detection can adopt frame difference method, optical flow method or background subtraction method.

[0077] Further, in the step S20, the mask can be manually extracted, or can be automatically extracted through image segmentation or target detection.

[0078] In the embodiment, since the method determines whether the coal or ore on the scraper chain is in a motion state based on the position of the coal or ore on the picture at different times, it is necessary to determine whether the first image received after starting is the first frame image. If it is the first frame image, only the corresponding processing of the first image is needed; if the first image is not the first frame image, not only the target on the image needs to be processed, but also the image needs to be compared with the previous frame image to determine whether the position of the target on the picture has moved.

[0079] The application also provides a visual detection device for chain breakage fault of a scraper conveyor, which comprises a controller configured to execute the program of the visual detection method provided in any of the above embodiments. Further, the device is composed of two parts, i.e., an underground part and an overground part. The underground part comprises an industrial camera, a lens arranged on the industrial camera and an embedded processing platform connected with the industrial camera. The overground part mainly comprises an industrial computer. Considering the limited data processing capacity of the embedded processing platform, a processing center is established in the overground part to directly process the video signals collected by the industrial camera.

[0080] In the embodiment, the industrial camera, the lens and the like are arranged at the head part, the middle part and the tail part of the scraper conveyor to obtain real-time video information of the scraper conveyor, and the video information is transmitted to the embedded platform to obtain real-time image data of the scraper conveyor.

[0081] As shown in Figure 4 The application also provides a step of transplanting the algorithm to the embedded processing platform of the Raspberry Pi:

[0082] Step S41: The system is burned into the embedded processing platform, and the system is started by connecting the display screen;

[0083] Step S42: The SSH service of the Raspberry Pi is started, the WIFI is connected and the IP address of the Raspberry Pi is obtained, the embedded processing platform is remotely logged in by means of the corresponding tool software, and the algorithm file is transplanted into the embedded processing platform;

[0084] Step S43: The camera is jointly debugged according to the camera model to determine whether it normally works and achieves the expected target;

[0085] Step S44: The self-starting on startup is set to facilitate the direct control of the user.

[0086] The technical solutions of the application have been described in combination with the preferred embodiments shown in the drawings, but it is easy for those skilled in the art to understand that the protection scope of the application is obviously not limited to these specific embodiments. Those skilled in the art can make equivalent changes or replacements to the related technical features without departing from the principles of the application, and the technical solutions after the changes or replacements will all fall within the protection scope of the application.

Claims

1. A visual detection method for chain breakage faults in scraper conveyors, characterized in that, The visual inspection method includes the following steps: Step S10: Acquire an image of the scraper conveyor; Step S20: Determine whether the image is empty. If it is, stop the program. If not, proceed to step S30. Step S30: Perform a first processing on the image to obtain a first image; the first processing includes performing grayscale processing, normalization processing and histogram equalization processing on the image in sequence, and the processed first image has the characteristics of uniform grayscale probability density distribution; Step S40: Determine whether the first image is the first frame. If yes, proceed to step S30; otherwise, proceed to step S50. Step S50: Perform a second processing on the first image to obtain a second image, and obtain the number n of moving objects in the second image; Step S60: Determine if n is zero. If yes, proceed to step S70; otherwise, proceed to step S30. Step S70: Obtain the sprocket speed signal of the scraper conveyor; Step S80: Determine if the sprocket speed is zero. If it is, determine that the machine is stopped normally. If not, execute the chain breakage alarm procedure.

2. The visual inspection method according to claim 1, characterized in that, The calculation formulas for the normalization and histogram equalization processes are as follows: s=T(r)=∫0p r (r)d r In the above formula, s represents the image pixel value after histogram equalization, r represents the image pixel value after normalization, T(r) represents the transformation function, and P r (r) represents the probability density function of the normalized image pixel value r.

3. The visual inspection method according to claim 1, characterized in that, In step S10, video information of the scraper conveyor is first acquired, and then the video information is processed to obtain a real-time image.

4. The visual inspection method according to claim 1, characterized in that, In step S30, the "second processing" includes the following steps: Step S31: Perform moving target detection to obtain a difference image or other moving target information; Step S32: Intersect the difference portion with the mask to obtain the scraper conveyor portion of the image, which is the first intermediate image; Step S33: Binarize the first intermediate image to obtain the second intermediate image; Step S34: Perform morphological processing on the second intermediate image to obtain the second image.

5. The visual inspection method according to claim 4, characterized in that, The binarization process is as follows: A suitable threshold T is selected, and each pixel is processed using the following formula; pixels with values ​​greater than or equal to the threshold T in the grayscale image are covered with 225, and the binarization result of these pixels is the foreground; pixels with values ​​less than the threshold T in the grayscale image are covered with 0, and the binarization result of these pixels is the background, thus obtaining the binarized image. In the above formula, p(x,y) represents the pixel value at pixel (x,y) after the grayscale image is binarized, and gray(x,y) represents the pixel value at pixel (x,y) of the grayscale image.

6. The visual inspection method according to claim 4, characterized in that, The morphological processing method is as follows: perform erosion and dilation operations on the second intermediate image sequentially.

7. The visual inspection method according to claim 4, characterized in that, The moving target detection employs inter-frame difference, optical flow, or background subtraction methods.

8. The visual inspection method according to claim 4, characterized in that, In step S20, the mask is obtained by manual extraction, or by automatic extraction through image segmentation or target detection.

9. A visual inspection device for chain breakage faults in scraper conveyors, the visual inspection device comprising a controller, characterized in that, The controller is configured to perform the visual inspection method according to any one of claims 1-8.

Images