Scraper chain plate speed detection method and device based on machine vision
Through machine vision technology, the scraper movement track area is identified and the chain plate speed is detected, which solves the problem of missing chain detection function of scraper equipment, real-time and accurate chain plate speed monitoring and chain break alarm, and improves the safety of the equipment Sex and reliability.
Patent Information
- Application Number
- CN202411240697.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-05
- Publication Date
- 2024-12-03
AI Technical Summary
The existing scraper equipment lacks chain speed monitoring and chain break detection functions, resulting in hidden dangers in the operation of the equipment, easy damage to the sensor, and manual inspections cannot be checked in real time.
Using machine vision-based detection method, through image acquisition, processing and display units, the scraper motion track area is identified, the background difference method and mask binary map are used to distinguish background and foreground, the contour area is detected, and the chain plate movement speed is calculated. If the speed is constant, a link break alarm signal will be sent.
Real-time and accurate detection of the speed of the scraper chain plate is realized, filling the gap in the industry, and improving equipment safety and reliability.
Smart Images

Figure CN119059210A_ABST
Abstract
Description
Technical Field
[0001] This patent application belongs to the field of scraper conveyor visual detection technology, and more specifically, it relates to a method and device for detecting the chain speed of a scraper conveyor based on machine vision. Background Art
[0002] Scrapers are used in the industrial and mining industries to transport materials, especially in coal washing processes. They are indispensable equipment. The main mechanism includes a certain number of chain plates connecting two chains. One end of the chain is driven by a driving mechanism. The chain plate moves with the chain in the groove, hanging the material and moving it forward to achieve the purpose of conveying the material. However, there may be problems with the chain breaking, which will force the equipment to stop operating.
[0003] The current detection method is to perform manual inspection and add sensors that can reflect certain state parameters of the scraper. Since the internal working conditions of the scraper are random and uncontrollable, the chain speed cannot be monitored, the sensor is easily damaged, and manual inspection cannot be carried out in real time, the existing scraper equipment does not have the function of detecting whether the chain is broken. Summary of the invention
[0004] The technical problem to be solved by the present invention is to provide a method and device for detecting the chain speed of a scraper conveyor based on machine vision, which can solve the problem of lack of chain speed monitoring and early warning of chain breakage in related technologies.
[0005] In order to solve the above problems, the technical solution adopted by the present invention is:
[0006] A method for detecting the chain speed of a scraper conveyor based on machine vision, the steps are as follows:
[0007] Step 1: Collect images perpendicular to the working surface of the scraper;
[0008] Step 2: Select a certain area of the scraper track in the image and identify the scraper in this area, which is referred to as the track area below. Figure 2 The area enclosed by the blue lines shown in the figure, and the length of this area in the track direction is less than the distance between the two scrapers, ensuring that only one scraper moves in the area; this operation requires segmenting the area in the image, and selecting any area in the image that covers the scraper track to identify the scraper;
[0009] Step 3: Distinguish the background and foreground in the image;
[0010] Step 4: Detect the mask of the moving result and output the foreground as a mask binary image, with the background displayed in black and the foreground displayed in white;
[0011] step5: contour detection;
[0012] Step 6: Determine the size of the contour area. If it is larger than the threshold (80% of the total number of pixels occupied by the scraper in the track area is used as the threshold), it will be retained and proceed to step 7; if it is smaller than the threshold, it will not be retained;
[0013] Step 7: Record the position coordinates, width and height of the circumscribed rectangle of the outline, and calculate the center coordinates of the circumscribed rectangle;
[0014] Step 8: Mark the id of the bounding rectangle of the outline, and before marking, check whether it has the same id as the bounding rectangle in the previous frame;
[0015] Step 9: Count the first and last positions in the same scraper ID set, calculate the displacement difference and time difference, and obtain the moving speed of the scraper in the image. If the speed is constant at the speed value of the previous scraper ID, proceed to step 10;
[0016] step10: Send out a chain break alarm signal.
[0017] Furthermore, in step 3, the background difference method is used to distinguish the background and foreground in the image. The background difference method is used to monitor the changing parts in the picture and generate a mask image. The scraper equipment has an upper chain plate and a lower chain plate in operation. During the cycle operation, the upper and lower chain plates will appear in the picture at the same time and are in motion, which will interfere with the recognition and make it impossible to track a specific chain plate. Therefore, only the analysis area is selected on the track to meet the conditions that the background is stationary and there is only a single moving object in the foreground.
[0018] Furthermore, in step 4, the mask binary image is opened to remove noise, so as to improve the recognition accuracy in the subsequent edge detection.
[0019] Furthermore, in step 6, the threshold is set by taking 80% of the number of pixels occupied by the scraper in the recognition area of the image as the threshold. According to the actual recognition effect, the threshold can be adjusted between 50% and 90% to ensure that the position of the scraper can be accurately tracked in the binary image of the foreground.
[0020] Furthermore, in step 8, the circumscribed rectangle of the outline is marked with an ID, so that it can be identified whether the foreground in different frame images is the same chain plate, and the timestamp when the ID is marked is recorded.
[0021] Furthermore, in step 8, the basis for "checking whether the id is the same as the bounding rectangle in the previous frame before marking" is: if the position distance between the two bounding rectangles is greater than the set pixel value, it is determined to be a new scraper id, otherwise it is the scraper id in the previous frame image, and the new coordinates and width and height values are updated to the scraper id of the previous frame.
[0022] A device for detecting the chain speed of a scraper conveyor based on machine vision utilizes the above method and includes an image acquisition unit, an image processing unit, and an image display unit. The image acquisition unit is used to execute step 1, the image processing unit is used to execute step 2 to step 10, and the image display unit is used to display the detection results.
[0023] Due to the adoption of the above technical solution, the beneficial effects achieved by the present invention are:
[0024] The present invention is simple, efficient and highly accurate, and can detect the chain plate speed in real time to determine whether the chain is in normal use, thus filling a gap in the industry. BRIEF DESCRIPTION OF THE DRAWINGS
[0025] Figure 1 It is a flow chart of the visual inspection method of the present invention;
[0026] Figure 2 is the first frame image processing result in the present invention;
[0027] Figure 3 This is the processing result of the second frame image in the present invention. DETAILED DESCRIPTION
[0028] The present invention is further described in detail below with reference to the embodiments.
[0029] A method for detecting the chain speed of a scraper conveyor based on machine vision, such as Figure 1 This embodiment proposes a method for detecting the speed of a scraper chain based on machine vision, comprising:
[0030] Step 1: The camera is perpendicular to the working surface of the scraper to collect images;
[0031] Step 2: Select a certain area of the scraper track in the image and identify the scraper in this area, such as Figure 2 The area enclosed by the blue lines shown in the figure, and the length of this area in the track direction is less than the distance between the two scrapers, ensuring that only one scraper moves in the area; this operation requires segmenting the area in the image, selecting any area in the image that covers the scraper track to identify the scraper, and the identified area can be referred to as the track area;
[0032] Step 3: Use background difference method to distinguish the background and foreground in the image;
[0033] The background difference method is used to monitor the changing parts in the picture and generate a mask image. When the scraper equipment is working, there are upper and lower chain plates. During the cycle operation, the upper and lower chain plates will appear in the picture at the same time and are both moving, which will interfere with the recognition and make it impossible to track a specific chain plate. Therefore, only the analysis area is selected on the track to meet the conditions of a static background and a single moving object in the foreground.
[0034] Step 4: Output the foreground as a masked binary image, the background (stationary track) is displayed in black, and the foreground (moving scraper) is displayed in white;
[0035] Step 5: Contour detection, detect the contour of the foreground;
[0036] Step 6: Determine the size of the contour area. If it is larger than the threshold (50% to 90% of the total number of pixels occupied by the scraper in the track area / recognition area is used as the threshold, such as 60% or 80%), it will be retained and proceed to step 7; if it is smaller than the threshold, it will not be retained. Filter out the identified noise points;
[0037] Step 7: Record the position coordinates, width and height of the circumscribed rectangle of the contour, and calculate the center coordinates of the circumscribed rectangle. For example, use the boundingRect() function of the cv library to calculate the coordinates of the four points of the circumscribed rectangle of the contour.
[0038] (x1,y1), (x2,y1), (x2,y2), (x1,y2) calculate the shape center coordinates (x01,y01)
[0039] =((x1+x2) / 2,(y1+y2) / 2);
[0040] Step 8: Mark the id of the circumscribed rectangle of the outline. Before marking, check whether it is the same id as the circumscribed rectangle in the previous frame. The basis is: if the position distance between the two circumscribed rectangles is greater than the set pixel value, it is determined to be a new scraper id, otherwise it is the scraper id in the previous frame image. The new coordinates and width and height values are updated to the scraper id of the previous frame.
[0041] For example, the center coordinates of the circumscribed rectangle of the current image outline (x01, y01) are marked with id1, and the center coordinates of the circumscribed rectangle of the second frame image are calculated after processing (x02, y02), and the Euclidean distance between the centers of the circumscribed rectangles of the two frames is calculated. Is it greater than the set pixel value threshold (p = (x2-x1) / 3), if dis is greater than or equal to p, it is determined that the new scraper is marked as id2, and it also means that the tracking of the center coordinates of the circumscribed rectangle marked as id1 (x01, y01) is completed. If dis is less than p, the center coordinates of the circumscribed rectangle (x02, y02) in the second frame image are replaced with the center coordinates of the circumscribed rectangle marked as id1, that is, the scraper in the two frames is the same scraper. The center coordinates of the circumscribed rectangle of the scraper in the second frame image are (x02, y02). The third frame image is processed in the same way to determine whether the center coordinates of the circumscribed rectangle (x03, y03) are the same scraper. And so on, until the scraper id is tracked to the right edge of the track area, the center coordinates of the circumscribed rectangle tracked with the same id are counted and performed step 9;
[0042] Step 9: Count the first and last positions in the same scraper ID set, calculate the displacement difference and time difference, and get the moving speed of the scraper in the image. If the speed is constant at the speed value of the previous scraper ID, proceed to step 10. Specifically, count the first and last positions in the same scraper circumscribed rectangle center coordinate set ((x01, y01), (x02, y02), (x03, y03), (x4, y04)... (x0n, y0n)), calculate the displacement difference The time taken to calculate the center of the circumscribed rectangle of the same scraper is counted, and the time difference ΔT between the first and the last one is calculated, and the moving speed of the scraper in the image V = ΔS / ΔT is obtained;
[0043] Step 10: Continuously count the speed V of each scraper. If the value of speed V changes suddenly, a chain break alarm signal will be issued in the image display unit.
[0044] In the above steps, select the track area in the image, such as Figure 2 The blue wireframe shown is used to identify the scraper position and perform speed calculation.
[0045] In the above steps, the morphological processing method is: performing an opening operation on the mask binary image to remove noise points, so as to improve the recognition accuracy in the subsequent edge detection.
[0046] In the above steps, it is possible to identify whether the foreground in different frame images is the same chain plate, set an ID tag, and record the time when the ID is marked.
[0047] For specific results, see Figure 2 As shown, the blue box is the track area selected for identification, the area surrounded by the red line is the foreground range after noise filtering, and the green box represents the circumscribed rectangle of the foreground. Figure 3 Yes Figure 2The tracking effect of the scraper can be seen from the ID value. The scraper is identified as the same scraper, and the speed of the scraper can be calculated by position and time.
[0048] The present invention also provides an integrated design structure solution, that is, a detection device, which includes an image acquisition unit, an image processing unit, and an image display unit, wherein the image acquisition unit and the image display unit are connected to the image processing unit. The image acquisition unit is used to execute step 1, the image processing unit is used to execute step 2 to step 10, and the image display unit is used to display the processing results such as Figure 2 , Figure 3 shown.
Claims
1. A method for detecting the speed of a scraper chain based on machine vision, characterized in that The steps include: Step 1: Collect images perpendicular to the working surface of the scraper; Step 2: Select a certain area of the scraper track in the image, identify the scraper in this area, and ensure that the length of this area in the track direction is less than the distance between the two scrapers to ensure that only one scraper moves in the area; Step 3: Distinguish the background and foreground in the image; Step 4: Detect the mask of the moving result and output the foreground as a mask binary image, with the background displayed in black and the foreground displayed in white; Step 5: Contour detection, detect the contour of the foreground; Step 6: Determine the size of the contour area. If it is larger than the threshold, keep it and proceed to step 7. If it is smaller than the threshold, do not keep it. Step 7: Record the position coordinates, width and height of the circumscribed rectangle of the outline, and calculate the center coordinates of the circumscribed rectangle; Step 8: Mark the id of the bounding rectangle of the outline, and before marking, check whether it has the same id as the bounding rectangle in the previous frame; Step 9: Count the first and last positions in the same scraper ID set, calculate the displacement difference and time difference, and obtain the moving speed of the scraper in the image. If the speed is constant at the speed value of the previous scraper ID, proceed to step 10; step10: Send out a chain break alarm signal.
2. According to the method for detecting the speed of a scraper chain based on machine vision according to claim 1, it is characterized in that: In step 3, the background difference method is used to distinguish the background and foreground in the image.
3. The method for detecting the speed of a scraper chain based on machine vision according to claim 1 is characterized in that: In step 4, the mask binary image is opened to remove noise, so as to improve the recognition accuracy in subsequent edge detection.
4. The method for detecting the speed of a scraper chain based on machine vision according to claim 1 is characterized in that: The threshold value in step 6 is set to 80% of the number of pixels occupied by the scraper in the recognition area of the image.
5. The method for detecting the speed of a scraper chain based on machine vision according to claim 1 is characterized in that: In step 8, the circumscribed rectangle of the outline is marked with an ID, which can identify whether the foreground in different frame images is the same chain plate, and record the timestamp when the ID is marked.
6. The method for detecting the speed of a scraper chain based on machine vision according to claim 5 is characterized in that: In step 8, the basis for "checking whether the id is the same as the bounding rectangle in the previous frame before marking" is: if the position distance between the two bounding rectangles is greater than the set pixel value, it is determined to be a new scraper id, otherwise it is the scraper id in the previous frame image, and the new coordinates and width and height values are updated to the scraper id of the previous frame.
7. A device for detecting the speed of a scraper chain based on machine vision, using the method described in any one of claims 1 to 6, characterized in that: It includes an image acquisition unit, an image processing unit, and an image display unit. The image acquisition unit and the image display unit are both connected to the image processing unit. The image acquisition unit is used to execute step 1, the image processing unit is used to execute step 2 to step 10, and the image display unit is used to display the detection results.
