A method for detecting chain breakage of a warehouse scraper based on binocular vision and line laser

Abstract

The present application proposes a method for detecting chain breakage of a warehouse scraper based on binocular vision and line laser, which relates to the technical field of scraper conveyors, wherein the method comprises: using a line laser as a light source, taking pictures of the warehouse scraper vertically downward through a binocular camera, obtaining a warehouse scraper image and a warehouse scraper bottom image; judging whether the warehouse scraper is started through the image taken by the binocular camera; if it is determined that the warehouse scraper is in a stationary state, using the warehouse scraper image taken by the binocular camera as the original point cloud data, judging whether the warehouse scraper has a chain break based on the original point cloud data through a segmentation algorithm and image processing method; if it is determined that the warehouse scraper is in a started state, using the warehouse scraper image and the warehouse scraper bottom image to judge whether the warehouse scraper has a chain break. The present invention adopting the above scheme realizes accurate detection of chain breakage faults.

Description

Technical Field

[0001] The present application relates to the technical field of scraper conveyors, and in particular to a method for detecting chain breakage of a bin-mounted scraper conveyor based on binocular vision and line laser. Background Art

[0002] With the increasing mechanization and automation of coal mines, scraper conveyors have become indispensable transportation equipment in coal mines and play an important role in the production process of coal mines. Chain breakage of scraper conveyors is the most common failure of scraper conveyors. Due to its various causes, the unpredictable location of the breakage, poor working environment conditions, and small and difficult fault handling space, it directly restricts the high-yield and efficient production needs of mines and seriously threatens the safety of mine production and the personal safety of employees. Summary of the invention

[0003] The present application aims to solve one of the technical problems in the related art at least to some extent.

[0004] To this end, the first purpose of this application is to propose a method for detecting broken chain of a silo scraper based on binocular vision and line laser, which solves the technical problem that the existing methods cannot detect broken chain failures in time, endangering the normal operation of the mine, and realizes accurate detection of broken chain failures.

[0005] The second purpose of the present application is to propose a device for detecting chain breakage of a bin-loading scraper conveyor based on binocular vision and line laser.

[0006] To achieve the above-mentioned objectives, the first embodiment of the present application proposes a method for detecting broken chains of a bin-loading scraper based on binocular vision and line laser, comprising: using a line laser as a light source, taking pictures of the bin-loading scraper vertically downward through a binocular camera, and obtaining an image of the bin-loading scraper and an image of the bottom of the bin-loading scraper; judging whether the bin-loading scraper is started through the image taken by the binocular camera; if it is determined that the bin-loading scraper is in a stationary state, using the image of the bin-loading scraper taken by the binocular camera as the original point cloud data, and judging whether the bin-loading scraper has a broken chain based on the original point cloud data through a segmentation algorithm and image processing method; if it is determined that the bin-loading scraper is in a started state, using the image of the bin-loading scraper and the image of the bottom of the bin-loading scraper to judge whether the bin-loading scraper has a broken chain.

[0007] The bin scraper chain break detection method based on binocular vision and line laser in the embodiment of the present application uses a line laser combined with a binocular camera to capture real-time images of the bin scraper, realizes real-time chain break fault detection of the bin scraper based on the captured image data, and promptly formulates corresponding preventive measures, which can greatly reduce the number of chain breaks and ensure the normal operation of the mine.

[0008] Optionally, in one embodiment of the present application, judging whether the warehouse scraper is started by using an image captured by a binocular camera includes:

[0009] Based on the image of the warehouse scraper taken by the binocular camera and the image of the bottom of the warehouse scraper machine, a first distance from the binocular camera to the warehouse scraper and a second distance from the binocular camera to the bottom of the warehouse scraper machine are calculated;

[0010] If the first distance and the second distance appear alternately within a certain period of time and the appearance time is stable, it is determined that the warehouse scraper is in the starting state;

[0011] If the first distance and the second distance do not appear alternately within a certain period of time or the alternating time is unstable, it is determined that the warehouse scraper is in a state of running to a standstill.

[0012] Optionally, in one embodiment of the present application, judging whether there is a chain break in the warehouse scraper conveyor based on the original point cloud data by segmentation algorithm and image processing method includes:

[0013] The original 3D point cloud data is mapped into a 2D depth image by projection;

[0014] Use the chain segmentation algorithm model to perform pixel-level positioning on the two-dimensional depth image to determine the overall outline of the chain;

[0015] The pixel values ​​inside and outside the overall outline of the chain of the two-dimensional depth image are set to convert the two-dimensional depth image into a binary image;

[0016] The binary image is processed by using corrosion operation and expansion operation to obtain a processed binary image;

[0017] The canny operator is used to extract the edge information of the processed binary image and determine the coordinates of the edge image pixels;

[0018] Use multiple edge pixel points of the edge image to fit a straight line in the x-axis direction. If the fitting is unsuccessful, it is judged that the warehouse scraper has no chain break. If two straight lines are fitted, it is judged that the warehouse scraper has a chain break.

[0019] Optionally, in one embodiment of the present application, using the warehouse scraper image and the warehouse scraper machine bottom image to determine whether the warehouse scraper machine has a broken chain includes:

[0020] If the binocular camera takes a picture of the warehouse scraper and fails to take a new scraper picture within a certain period of time, it is judged that the scraper double chain is broken behind the scraper;

[0021] If the bin-loading scraper is not captured by the bin-loading scraper after the bin-loading scraper bottom image is captured by the bin-loading camera within a certain period of time, it is determined that the scraper double chain is broken in front of the scraper.

[0022] Optionally, in one embodiment of the present application, using the warehouse scraper image and the warehouse scraper machine bottom image to determine whether the warehouse scraper machine has a broken chain also includes:

[0023] After the binocular camera captures the image of the bin-mounted scraper, the bin-mounted scraper image and the bin-mounted scraper machine image are continuously captured within a certain period of time, and the distance from the binocular camera to the bin-mounted scraper and the bottom of the bin-mounted scraper machine is measured based on the captured images to obtain a set of the first pixel points corresponding to the bin-mounted scraper image and the second pixel points corresponding to the bottom image of the bin-mounted scraper machine;

[0024] If the first pixel number and the corresponding second pixel number in the set meet the preset condition, the position point corresponding to the bin scraper is marked, and a straight line is generated based on the marked multiple position points;

[0025] The tilt angle of the warehouse scraper is calculated based on the generated straight line. If the angle is greater than the preset threshold, it is judged that the single chain of the warehouse scraper is broken. If the angle is less than the preset threshold, it is judged that a skew chain appears on the warehouse scraper.

[0026] To achieve the above-mentioned purpose, the second embodiment of the present invention proposes a chain break detection device for a warehouse scraper based on binocular vision and line laser, including an image acquisition module, a first judgment module, a second judgment module, and a third judgment module, wherein:

[0027] An image acquisition module is used to use a line laser as a light source to take pictures of the bin scraper vertically downward through a binocular camera to obtain images of the bin scraper and the bottom of the bin scraper;

[0028] The first judgment module is used to judge whether the warehouse scraper is started through the image taken by the binocular camera;

[0029] The second judgment module is used to determine whether the warehouse scraper is broken by using the warehouse scraper image taken by the binocular camera as the original point cloud data if it is determined that the warehouse scraper is in a stationary state, and based on the original point cloud data, through a segmentation algorithm and image processing method;

[0030] The third judgment module is used to judge whether there is a chain break in the bin-matching scraper by using the bin-matching scraper image and the bin-matching scraper bottom image if it is determined that the bin-matching scraper is in the startup state.

[0031] Optionally, in one embodiment of the present application, judging whether the warehouse scraper is started by using an image captured by a binocular camera includes:

[0032] Based on the image of the warehouse scraper taken by the binocular camera and the image of the bottom of the warehouse scraper machine, a first distance from the binocular camera to the warehouse scraper and a second distance from the binocular camera to the bottom of the warehouse scraper machine are calculated;

[0033] If the first distance and the second distance appear alternately within a certain period of time and the appearance time is stable, it is determined that the warehouse scraper is in the starting state;

[0034] If the first distance and the second distance do not appear alternately within a certain period of time or the alternating time is unstable, it is determined that the warehouse scraper is in a state of running to a standstill.

[0035] Optionally, in one embodiment of the present application, judging whether there is a chain break in the warehouse scraper conveyor based on the original point cloud data by segmentation algorithm and image processing method includes:

[0036] The original 3D point cloud data is mapped into a 2D depth image by projection;

[0037] Use the chain segmentation algorithm model to perform pixel-level positioning on the two-dimensional depth image to determine the overall outline of the chain;

[0038] The pixel values ​​inside and outside the overall outline of the chain of the two-dimensional depth image are set to convert the two-dimensional depth image into a binary image;

[0039] The binary image is processed by using corrosion operation and expansion operation to obtain a processed binary image;

[0040] The canny operator is used to extract the edge information of the processed binary image and determine the coordinates of the edge image pixels;

[0041] Use multiple edge pixel points of the edge image to fit a straight line in the x-axis direction. If the fitting is unsuccessful, it is judged that the warehouse scraper has no chain break. If two straight lines are fitted, it is judged that the warehouse scraper has a chain break.

[0042] Optionally, in one embodiment of the present application, using the warehouse scraper image and the warehouse scraper machine bottom image to determine whether the warehouse scraper machine has a broken chain includes:

[0043] If the binocular camera takes a picture of the warehouse scraper and fails to take a new scraper picture within a certain period of time, it is judged that the scraper double chain is broken behind this scraper;

[0044] If the bin-loading scraper is not captured by the bin-loading scraper after the bin-loading scraper bottom image is captured by the bin-loading camera within a certain period of time, it is determined that the scraper double chain is broken in front of the scraper.

[0045] Optionally, in one embodiment of the present application, using the warehouse scraper image and the warehouse scraper machine bottom image to determine whether the warehouse scraper machine has a broken chain also includes:

[0046] After the binocular camera captures the image of the bin-mounted scraper, the bin-mounted scraper image and the bin-mounted scraper machine image are continuously captured within a certain period of time, and the distance from the binocular camera to the bin-mounted scraper and the bottom of the bin-mounted scraper machine is measured based on the captured images to obtain a set of the first pixel points corresponding to the bin-mounted scraper image and the second pixel points corresponding to the bottom image of the bin-mounted scraper machine;

[0047] If the first pixel number and the corresponding second pixel number in the set meet the preset condition, the position point corresponding to the bin scraper is marked, and a straight line is generated based on the marked multiple position points;

[0048] The tilt angle of the warehouse scraper is calculated based on the generated straight line. If the angle is greater than the preset threshold, it is judged that the single chain of the warehouse scraper is broken. If the angle is less than the preset threshold, it is judged that a skew chain appears on the warehouse scraper.

[0049] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0051] Figure 1 A schematic flow chart of a method for detecting chain breakage of a warehouse scraper based on binocular vision and line laser provided in Example 1 of the present application;

[0052] Figure 2 This is an example diagram of a bin-scraping machine according to an embodiment of the present application;

[0053] Figure 3 This is a flow chart for determining whether a warehouse scraper has a broken chain when the scraper is stationary in an embodiment of the present application;

[0054] Figure 4 A schematic structural diagram of a chain break detection device for a bin-loading scraper conveyor based on binocular vision and line laser provided in an embodiment of the present application. DETAILED DESCRIPTION

[0055] Embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0056] The following describes the method and device for detecting chain breakage of a bin-loading scraper conveyor based on binocular vision and line laser in an embodiment of the present application with reference to the accompanying drawings.

[0057] Figure 1 A flow chart of a method for detecting chain breakage of a bin-loading scraper conveyor based on binocular vision and line laser provided in Example 1 of the present application.

[0058] like Figure 1 As shown, the bin scraper chain break detection method based on binocular vision and line laser includes the following steps:

[0059] Step 101, using a line laser as a light source, taking pictures of the warehouse scraper conveyor vertically downward through a binocular camera to obtain images of the warehouse scraper conveyor and the bottom of the warehouse scraper conveyor;

[0060] In this embodiment, the equipment used includes a binocular camera and a line laser system S1. The binocular camera is installed on a gantry or a roof to shoot the warehouse scraper vertically downward. The camera lens is 1.5 meters away from the warehouse scraper, and the warehouse scraper is on the ground. The reason for using the laser is that it is almost unaffected by the intensity of sunlight. The line laser is small in size, easy to install, and can measure distance.

[0061] Step 102, judging whether the warehouse scraper is started by using the image captured by the binocular camera;

[0062] In this embodiment, Figure 2 This is an example of a scraper conveyor. Figure 2 Before starting the warehouse scraper shown in the figure, it is necessary to obtain equipment information as prior information, including:

[0063] First measure the distance H meters from the binocular camera lens to the bottom of the bin scraper, and then measure the distance h meters from the camera lens to the scraper of the bin scraper;

[0064] Measure the scraper length of the warehouse scraper machine, use a binocular camera to pre-photograph the scraper, and obtain the number of pixels corresponding to the scraper image as M;

[0065] Measure the distance between two adjacent scrapers of the warehouse scraper machine is l meter, and the scraper running speed is vm / s (1.2 meters / second). Then the time from the camera to the two scrapers is t (seconds) = l / v;

[0066] After obtaining the prior information, the process of determining the start-up includes:

[0067] 1. Before the warehouse scraper is put into operation, the inspection worker will conduct an overall inspection of the scraper to ensure that there is no problem before starting the scraper;

[0068] 2. Start the camera. The camera automatically determines that the scraper is started by measuring the distance H and h alternately when the camera takes pictures of the scraper within a certain period of time (time greater than t);

[0069] 3. The camera automatically determines that the scraper has stopped by measuring the distance unchanged or the distance H and h appear alternately for a period of time greater than t when the camera shoots the scraper within a certain period of time (time greater than t), indicating that the scraper is slowly stopping from running.

[0070] 4. The scraper start and stop signals will also be connected to the camera.

[0071] Step 103, if it is determined that the warehouse scraper is in a stationary state, the warehouse scraper image taken by the binocular camera is used as the original point cloud data, and based on the original point cloud data, a segmentation algorithm and an image processing method are used to determine whether the warehouse scraper has a broken chain;

[0072] In this embodiment, the state of the bin scraper is judged as a priori condition when the bin scraper is stationary, specifically:

[0073] 1. When the scraper is stationary, use a binocular camera to take a picture of the scraper and mark the position of the chain in the image.

[0074] The binocular plus line laser method collects point cloud data. The overall idea is to obtain a two-dimensional depth image from the three-dimensional point cloud data through projection of the original point cloud data of the chain, and then locate the chain through a deep learning segmentation algorithm to determine the position, and use image processing to detect whether the chain has cracks, and issue an alarm according to the size of the crack.

[0075] Figure 3 This is a flow chart for determining whether there is a chain break in the warehouse scraper when the scraper is stationary in an embodiment of the present application. Figure 3 As shown, the process is as follows:

[0076] 1. The point cloud data is filtered through straight-through filtering and statistical outlier removal algorithms to filter out noise points.

[0077] 2. Map the 3D point cloud data to the 2D depth image by projection (additive formula can be used), that is, the mapping relationship between the point cloud and the image converts the 3D point (x, y, z) into 2D pixel points (u, v), and then draws the image based on the 2D pixel points.

[0078] Assuming there is a three-dimensional point (x, y,) in the camera coordinate system, its projection coordinates (u, v) on the image plane can be calculated by the following formula:

[0079] u=(fx*x+cx*z) / z

[0080] v=(f y *y+cy*z) / z

[0081] Among them, fx, fy, cx, and cy represent the internal parameters of the camera;

[0082] 3. Annotate the chain samples at the pixel level and use the segmentation algorithm to train the chain segmentation algorithm model, determine the pixel-level positioning of the chain, and determine the overall outline of the chain;

[0083] 4. Set the pixels inside the chain outline to 255 and the pixels in the rest of the chain to 0, and convert it into a binary image;

[0084] 5. After corrosion operation and expansion algorithm, the accuracy of edge extraction is guaranteed;

[0085] 6. Use the canny operator to extract image edge information and determine the coordinates of edge image pixels;

[0086] 7. Use multiple edge pixel points of the edge image to fit a straight line in the x-axis direction. If there is no straight line, it means there is no broken link. If there are multiple pixel points on a straight line, and two straight lines are fitted to meet this requirement, then calculate the distance D between the two straight lines. The alarm level is determined according to the size of D, which is divided into mild and severe warnings. When D is too large, there is a risk of broken link, and the platform is notified that a shutdown warning is required.

[0087] Step 104, if it is determined that the warehouse scraper is in the startup state, use the warehouse scraper image and the warehouse scraper bottom image to determine whether the warehouse scraper has a broken chain.

[0088] In this embodiment, when a chain break is detected, a local alarm sounds, the scraper stops running, and the platform is reported to notify relevant personnel to replace and repair it in time to avoid causing safety accidents.

[0089] In this embodiment, the distance can be measured by a binocular laser camera. From the prior information, it can be known that when the camera collects an image at a distance of h meters, it is a scraper image, otherwise when the distance is H meters, it is a bottom image of the warehouse scraper;

[0090] The warehouse scraper starts running, the camera starts working, the scraper stops, and the camera stops working.

[0091] Based on the data collected by the camera, the status of the warehouse scraper is divided into the following five situations. Specifically,

[0092] Case 1: When the camera captures the first scraper image, the timing t1 starts. If the second scraper image is not captured at time t+t1, it is considered that the scraper double chain breaks behind this scraper.

[0093] Case 2: When the camera captures the first scraper image, and can only capture this scraper afterwards, that is, it cannot capture the bottom image of the warehouse scraper, it is considered that the scraper double chain breaks in front of this scraper;

[0094] Case 3: When the camera captures the image of the bottom of the scraper conveyor and can continuously capture the image of the bottom of the scraper conveyor afterwards, that is, the scraper of the bin-feeding scraper conveyor cannot be captured, it is considered that there is a break in the double-chain of the scraper in front of this scraper;

[0095] Case 4: When a single-chain break occurs, the scraper on the broken-chain side will tilt backward. For example, when the right chain breaks, after the camera captures the image of the first scraper and after a time t, by measuring the distances from the camera to the scraper and the bottom of the bin-feeding scraper conveyor, the number of pixels corresponding to the scraper image is obtained as h1, h2…hm, with a total of m, and the number of pixels corresponding to the image of the bottom of the bin-feeding scraper conveyor is H1, H2…Hn, with a total of n. When m < M / 3 and n < M, it is marked as point a1, and as the scraper conveyor runs, m < M / 3 and n < M will last for a period of time until no m appears, and then it will be continuously marked as a2, a3, a4…aq, with a total of q points. Draw a straight line through these points and calculate the inclination angle φ of the scraper. When φ is large, it means that the inclination of the scraper is large. When φ is greater than 30 degrees, it is considered that there is a chain break in this scraper, stop the operation of the scraper, and report it to the platform to notify relevant personnel to replace and repair it in time to avoid safety accidents; when φ is small, it is considered that there is a slanted chain, and the upper computer platform side is notified for general alarm.

[0096] Case 5: Give an early warning when there are large cracks in the chain links, but there is no chain break at this time.

[0097] The method for detecting chain breaks in a bin-feeding scraper conveyor based on binocular vision and line laser proposed in this application uses a line laser combined with a binocular camera to take real-time images of the bin-feeding scraper conveyor, realizes real-time chain break fault detection of the bin-feeding scraper conveyor based on the captured image data, and formulates corresponding preventive measures in time, which can greatly reduce the number of chain breaks and ensure the normal operation of the mine.

[0098] To implement the above embodiments, this application also proposes a device for detecting chain breaks in a bin-feeding scraper conveyor based on binocular vision and line laser.

[0099] Figure 4 It is a schematic structural diagram of a device for detecting chain breaks in a bin-feeding scraper conveyor based on binocular vision and line laser provided in an embodiment of this application.

[0100] As Figure 4 shown, the device for detecting chain breaks in a bin-feeding scraper conveyor based on binocular vision and line laser includes an image acquisition module, a first judgment module, a second judgment module, and a third judgment module, where,

[0101] The image acquisition module is used to use the line laser as a light source and take pictures of the bin-feeding scraper conveyor vertically downward through the binocular camera to obtain the bin-feeding scraper image and the image of the bottom of the bin-feeding scraper conveyor;

[0102] The first judgment module is used to judge whether the warehouse scraper is started through the image taken by the binocular camera;

[0103] The second judgment module is used to determine whether the warehouse scraper is broken by using the warehouse scraper image taken by the binocular camera as the original point cloud data if it is determined that the warehouse scraper is in a stationary state, and based on the original point cloud data, through a segmentation algorithm and image processing method;

[0104] The third judgment module is used to judge whether there is a chain break in the bin-matching scraper by using the bin-matching scraper image and the bin-matching scraper bottom image if it is determined that the bin-matching scraper is in the startup state.

[0105] Optionally, in one embodiment of the present application, judging whether the warehouse scraper is started by using an image captured by a binocular camera includes:

[0106] Based on the image of the warehouse scraper taken by the binocular camera and the image of the bottom of the warehouse scraper machine, a first distance from the binocular camera to the warehouse scraper and a second distance from the binocular camera to the bottom of the warehouse scraper machine are calculated;

[0107] If the first distance and the second distance appear alternately within a certain period of time and the appearance time is stable, it is determined that the warehouse scraper is in the starting state;

[0108] If the first distance and the second distance do not appear alternately within a certain period of time or the alternating time is unstable, it is determined that the warehouse scraper is in a state of running to a standstill.

[0109] Optionally, in one embodiment of the present application, judging whether there is a chain break in the warehouse scraper conveyor based on the original point cloud data by segmentation algorithm and image processing method includes:

[0110] The original 3D point cloud data is mapped into a 2D depth image by projection;

[0111] Use the chain segmentation algorithm model to perform pixel-level positioning on the two-dimensional depth image to determine the overall outline of the chain;

[0112] The pixel values ​​inside and outside the overall outline of the chain of the two-dimensional depth image are set to convert the two-dimensional depth image into a binary image;

[0113] The binary image is processed by using corrosion operation and expansion operation to obtain a processed binary image;

[0114] The canny operator is used to extract the edge information of the processed binary image and determine the coordinates of the edge image pixels;

[0115] Use multiple edge pixel points of the edge image to fit a straight line in the x-axis direction. If the fitting is unsuccessful, it is judged that the warehouse scraper has no chain break. If two straight lines are fitted, it is judged that the warehouse scraper has a chain break.

[0116] Optionally, in one embodiment of the present application, using the warehouse scraper image and the warehouse scraper machine bottom image to determine whether the warehouse scraper machine has a broken chain includes:

[0117] If the binocular camera takes a picture of the warehouse scraper and fails to take a new scraper picture within a certain period of time, it is judged that the scraper double chain is broken behind this scraper;

[0118] If the bin-loading scraper is not captured by the bin-loading scraper after the bin-loading scraper bottom image is captured by the bin-loading camera within a certain period of time, it is determined that the scraper double chain is broken in front of the scraper.

[0119] Optionally, in one embodiment of the present application, using the warehouse scraper image and the warehouse scraper machine bottom image to determine whether the warehouse scraper machine has a broken chain also includes:

[0120] After the binocular camera captures the image of the bin-mounted scraper, the bin-mounted scraper image and the bin-mounted scraper machine image are continuously captured within a certain period of time, and the distance from the binocular camera to the bin-mounted scraper and the bottom of the bin-mounted scraper machine is measured based on the captured images to obtain a set of the first pixel points corresponding to the bin-mounted scraper image and the second pixel points corresponding to the bottom image of the bin-mounted scraper machine;

[0121] If the first pixel number and the corresponding second pixel number in the set meet the preset condition, the position point corresponding to the bin scraper is marked, and a straight line is generated based on the marked multiple position points;

[0122] The tilt angle of the warehouse scraper is calculated based on the generated straight line. If the angle is greater than the preset threshold, it is judged that the single chain of the warehouse scraper is broken. If the angle is less than the preset threshold, it is judged that a skew chain appears on the warehouse scraper.

[0123] It should be noted that the aforementioned explanation of the embodiment of the method for detecting chain breakage of a bin-mounted scraper conveyor based on binocular vision and line laser is also applicable to the device for detecting chain breakage of a bin-mounted scraper conveyor based on binocular vision and line laser in this embodiment, and will not be repeated here.

Claims

1. A method for detecting chain breakage of a warehouse scraper based on binocular vision and line laser, characterized in that: The following steps are involved: Using a line laser as a light source, a binocular camera is used to take pictures of the bin scraper machine vertically downward to obtain images of the bin scraper and the bottom of the bin scraper machine; The image captured by the bin-mounted camera is used to determine whether the warehouse scraper is started; If it is determined that the warehouse scraper is in a stationary state, the warehouse scraper image taken by the binocular camera is used as the original point cloud data, and based on the original point cloud data, a segmentation algorithm and image processing method are used to determine whether the warehouse scraper is broken; If it is determined that the warehouse scraper is in the starting state, use the warehouse scraper image and the warehouse scraper bottom image to determine whether the warehouse scraper has a broken chain; The method of judging whether there is a chain break in the warehouse scraper based on the original point cloud data by segmentation algorithm and image processing includes: The original 3D point cloud data is mapped into a 2D depth image by projection; Perform pixel-level positioning on the two-dimensional depth image using a chain segmentation algorithm model to determine the overall outline of the chain; Setting pixel values ​​inside and outside the overall outline of the chain of the two-dimensional depth image to convert the two-dimensional depth image into a binary image; Processing the binary image using corrosion operation and dilation operation to obtain a processed binary image; Using a canny operator to extract edge information of the processed binary image and determine the coordinates of edge image pixel points; Use multiple edge pixel points of the edge image to fit a straight line in the x-axis direction. If the fitting is unsuccessful, it is determined that the warehouse scraper has no chain break. If two straight lines are fitted, it is determined that the warehouse scraper has a chain break. The method of using the warehouse scraper image and the warehouse scraper machine bottom image to determine whether the warehouse scraper machine has a broken chain includes: If the binocular camera takes a picture of the warehouse scraper and fails to take a new scraper picture within a certain period of time, it is judged that the scraper double chain is broken behind the scraper; If the bin-loading scraper image is not captured after the bin-loading scraper image is captured by the bin-loading camera within a certain period of time, it is judged that the scraper double chain is broken in front of the scraper; The method of using the warehouse scraper image and the warehouse scraper bottom image to determine whether the warehouse scraper has a broken chain also includes: After the binocular camera captures the image of the bin-mounted scraper, the bin-mounted scraper image and the bin-mounted scraper machine image are continuously captured within a certain period of time, and the distance from the binocular camera to the bin-mounted scraper and the bottom of the bin-mounted scraper machine is measured based on the captured images to obtain a set of the first pixel points corresponding to the bin-mounted scraper image and the second pixel points corresponding to the bottom image of the bin-mounted scraper machine; If the first pixel number and the corresponding second pixel number in the set meet the preset condition, the position point corresponding to the bin scraper is marked, and a straight line is generated based on the marked multiple position points; The tilt angle of the warehouse scraper is calculated based on the generated straight line. If the angle is greater than the preset threshold, it is judged that the single chain of the warehouse scraper is broken. If the angle is less than the preset threshold, it is judged that a skew chain appears on the warehouse scraper.

2. The method according to claim 1, characterized in that The method of judging whether the warehouse scraper is started by using the image captured by the binocular camera includes: Based on the image of the warehouse scraper taken by the binocular camera and the image of the bottom of the warehouse scraper machine, a first distance from the binocular camera to the warehouse scraper and a second distance from the binocular camera to the bottom of the warehouse scraper machine are calculated; If the first distance and the second distance appear alternately within a certain period of time and the appearance time is stable, it is determined that the warehouse scraper is in the starting state; If the first distance and the second distance do not appear alternately within a certain period of time or the alternating time is unstable, it is determined that the warehouse scraper is in a state of running to a standstill.

3. A device for detecting chain breakage of a warehouse scraper based on binocular vision and line laser, characterized in that: It includes an image acquisition module, a first judgment module, a second judgment module, and a third judgment module, wherein: The image acquisition module is used to use the line laser as a light source to take pictures of the warehouse scraper vertically downward through a binocular camera to obtain images of the warehouse scraper and the bottom of the warehouse scraper; The first judgment module is used to judge whether the warehouse scraper is started through the image taken by the binocular camera; The second judgment module is used to determine whether the warehouse scraper is broken by using the warehouse scraper image taken by the binocular camera as the original point cloud data if it is determined that the warehouse scraper is in a stationary state, and based on the original point cloud data, through a segmentation algorithm and image processing method; The third judgment module is used to determine whether the warehouse scraper is broken by using the warehouse scraper image and the warehouse scraper bottom image if it is determined that the warehouse scraper is in the startup state; The method of judging whether there is a chain break in the warehouse scraper based on the original point cloud data by segmentation algorithm and image processing includes: The original 3D point cloud data is mapped into a 2D depth image by projection; Perform pixel-level positioning on the two-dimensional depth image using a chain segmentation algorithm model to determine the overall outline of the chain; Setting pixel values ​​inside and outside the overall outline of the chain of the two-dimensional depth image to convert the two-dimensional depth image into a binary image; Processing the binary image using corrosion operation and dilation operation to obtain a processed binary image; Using a canny operator to extract edge information of the processed binary image and determine the coordinates of edge image pixel points; Use multiple edge pixel points of the edge image to fit a straight line in the x-axis direction. If the fitting is unsuccessful, it is determined that the warehouse scraper has no chain break. If two straight lines are fitted, it is determined that the warehouse scraper has a chain break. The method of using the warehouse scraper image and the warehouse scraper machine bottom image to determine whether the warehouse scraper machine has a broken chain includes: If the binocular camera takes a picture of the warehouse scraper and fails to take a new scraper picture within a certain period of time, it is judged that the scraper double chain is broken behind the scraper; If the bin-loading scraper image is not captured after the bin-loading scraper image is captured by the bin-loading camera within a certain period of time, it is judged that the scraper double chain is broken in front of the scraper; The method of using the warehouse scraper image and the warehouse scraper machine bottom image to determine whether the warehouse scraper machine has a broken chain includes: If the binocular camera takes a picture of the warehouse scraper and fails to take a new scraper picture within a certain period of time, it is judged that the scraper double chain is broken behind this scraper; If the bin-loading scraper image is not captured after the bin-loading scraper image is captured by the bin-loading camera within a certain period of time, it is judged that the scraper double chain is broken in front of the scraper; The method of using the warehouse scraper image and the warehouse scraper bottom image to determine whether the warehouse scraper has a broken chain also includes: After the binocular camera captures the image of the bin-mounted scraper, the bin-mounted scraper image and the bin-mounted scraper machine image are continuously captured within a certain period of time, and the distance from the binocular camera to the bin-mounted scraper and the bottom of the bin-mounted scraper machine is measured based on the captured images to obtain a set of the first pixel points corresponding to the bin-mounted scraper image and the second pixel points corresponding to the bottom image of the bin-mounted scraper machine; If the first pixel number and the corresponding second pixel number in the set meet the preset condition, the position point corresponding to the bin scraper is marked, and a straight line is generated based on the marked multiple position points; The tilt angle of the warehouse scraper is calculated based on the generated straight line. If the angle is greater than the preset threshold, it is judged that the single chain of the warehouse scraper is broken. If the angle is less than the preset threshold, it is judged that a skew chain appears on the warehouse scraper.

4. The device according to claim 3, characterized in that The method of judging whether the warehouse scraper is started by using the image captured by the binocular camera includes: Based on the image of the warehouse scraper taken by the binocular camera and the image of the bottom of the warehouse scraper machine, a first distance from the binocular camera to the warehouse scraper and a second distance from the binocular camera to the bottom of the warehouse scraper machine are calculated; If the first distance and the second distance appear alternately within a certain period of time and the appearance time is stable, it is determined that the warehouse scraper is in the starting state; If the first distance and the second distance do not appear alternately within a certain period of time or the alternating time is unstable, it is determined that the warehouse scraper is in a state of running to a standstill.

Images