Visual Method, System, Program and Speed Regulation System for Measuring Coal Quantity on a Conveyor Belt

Abstract

The present invention relates to a visual method, system, program and speed control system for measuring the coal quantity on a conveyor belt. The visual method includes: collecting three-dimensional point cloud images of the conveyor belt by using upper and lower point cloud cameras; simplifying the three-dimensional point cloud images; judging whether there is coal on the current conveyor belt; selecting distances along the forward direction of the conveyor belt to extract the upper and lower point clouds and dividing the observation area thereof into grids, and performing random downsampling of the point clouds for each grid; fitting the upper and lower point clouds in each grid to obtain the upper and lower surface parameter equations; dividing the observation area of the coal quantity point cloud model into grids, calculating the volume through the area of each grid and the corresponding upper and lower surface parameter equations and summing them to obtain the passing volume of coal on the current conveyor belt; and repeating the foregoing steps, and accumulating the passing volume each time to obtain the passed coal quantity. The present invention can accurately measure the coal quantity on the conveyor belt of the conveyor, and is not affected by the change of the intensity of the underground ambient light, so as to achieve the effect of green energy-saving transportation of the conveyor.

Description

Technical Field

[0002] Generally speaking, the present invention relates to the technical field of coal conveying equipment; specifically, the present invention relates to a visual method, system, program and speed control system for measuring the coal quantity on a conveyor belt.

Background Art

[0004] In the coal production industry, belt conveyors are usually used for coal transportation. At present, the power of belt conveyors is mainly determined according to the peak production capacity of coal mines combined with corresponding redundancy factors, and they often run at a constant high speed after startup. Since the production capacity of the mining face is not uniform, it often leads to the problem of "using a big horse to pull a small cart" where the belt conveyor runs at a high speed with a low load or even no load, resulting in a large amount of wasted electric energy.

[0005] Therefore, there is an urgent need to propose a technical solution to solve the above problems.

Summary of the Invention

[0007] In view of this, the purpose of the present invention is to provide a visual method, system, program and speed control system for measuring the coal quantity on a conveyor belt, thereby solving or at least alleviating one or more of the above problems and other problems existing in the prior art.

[0008] To achieve the foregoing purpose, the first aspect of the present invention provides a visual method for measuring the coal quantity on a conveyor belt of a belt conveyor, wherein the visual method includes the following steps executed sequentially:

[0009] Step I: Use an upper 3D binocular point cloud camera and a lower 3D binocular point cloud camera to respectively collect a first three-dimensional point cloud image above and a second three-dimensional point cloud image below at a specific position of the conveyor belt;

[0010] Step II: Simplify the point cloud quantity of the first three-dimensional point cloud image and the second three-dimensional point cloud image by using the method of random downsampling;

[0011] Step III: Based on the simplified first three-dimensional point cloud image and the simplified second three-dimensional point cloud image, establish a coal quantity point cloud model of the coal quantity on the conveyor belt, compare the coal quantity point cloud model with the conveyor belt idling no-coal point cloud model, and judge whether there is coal on the current conveyor belt. If so, proceed to Step IV; otherwise, return to Step I;

[0012] Step IV: Select a section of distance along the forward direction of the conveyor belt to extract the first point cloud above the simplified conveyor belt, and select the same section of distance at the corresponding position along the forward direction of the conveyor belt to extract the second point cloud below the simplified conveyor belt;

[0013] Step V: dividing the observation area corresponding to the first point cloud and the second point cloud into point cloud grids, the number of squares obtained by the division is m, and then randomly downsampling the point clouds for each square of the first point cloud and the second point cloud, and storing the first point cloud and the second point cloud after the point cloud random downsampling;

[0014] Step VI: performing surface fitting on the first point cloud above the conveyor belt in each square to obtain an upper surface parameter equation f1, and performing surface fitting on the second point cloud below the conveyor belt in each square to obtain a lower surface parameter equation f2;

[0015] Step VII: Transform the upper surface parameter equation f1 into m point cloud points a1, a2, ..., a m It means that the lower surface parameter equation f2 is composed of m point cloud points b1, b2, ..., b m Indicates that the m point cloud points are spatial coordinates, and then a1, a2, ..., a m The horizontal position of b1, b2, ..., b m The horizontal positions of the m point cloud points correspond one to one, and the observation area of ​​the coal quantity point cloud model is divided into m uniform squares with the m point cloud points as spatial coordinates. The volume is calculated and summed by the area ds of each square and the upper surface parameter equation f1 and the lower surface parameter equation f2 at the corresponding square to obtain the current volume of coal on the conveyor belt. ;as well as

[0016] Step VIII: Repeat steps I to VII, and add up the passing volume each time to obtain the passing amount of coal.

[0017] Optionally, in the visual method as described above, in step VI, the process of fitting the upper surface parameter equation f1 is:

[0018] According to Y = AB, where A is [X1, X2, ..., X n ] T , B is [b0, b1, b2, b3, b4], Y is ln(z i ), X i is [1,x i , y i ,x i ², y i ²], B is the coefficient to be solved, where x i ,y i , z i is the coordinate of the i-th point cloud in the current grid, solve ;

[0019] Then the simultaneous equations

[0020]

[0021]

[0022]

[0023]

[0024]

[0025] It is solved that x0, y0, and z0 are

[0026]

[0027]

[0028] ,

[0029] Select x0, y0, and z0 of each grid, and then denote the x0, y0, and z0 calculated from the m grids as a i Store it as the parametric equation f1 representing the coal surface recorded by the upper 3D binocular point cloud camera;

[0030] In the step VI, the process of surface fitting the lower surface parametric equation f2 is as follows:

[0031] For the point cloud formed by the lower 3D binocular point cloud camera, the optimal solutions are x1, y1, z1. Select the closest point cloud x1, y1, z1 corresponding to the x0 and y0 obtained by using the process of surface fitting the upper surface parametric equation f1, where x1 and y1 are , and then denote the x0, y0, and z0 calculated from the m grids as b i And store it as the parametric equation f2 representing the coal surface recorded by the lower 3D binocular point cloud camera.

[0032] Optionally, in the visual method as described above, m is 100 of 10*10 or 128 of 16*8.

[0033] Optionally, in the visual method as described above, during the measurement process, a surface light source is used to provide a uniform lighting environment for the upper 3D binocular point cloud camera and the lower 3D binocular point cloud camera. The upper 3D binocular point cloud camera and the lower 3D binocular point cloud camera use the diffuse reflection of the surface of the coal and the conveyor belt through the surface light source to obtain the three-dimensional point cloud data of the coal.

[0034] Optionally, in the visual method as described above, the surface light source is an active light source that projects structured light.

[0035] Optionally, in the visual method as described above, in step II, the simplification reduces the number of point clouds in the first 3D point cloud image and the second 3D point cloud image to 1 / 4 of the original size, and in step V, the downsampling reduces the amount of point clouds to 1 / 4 of the original size but at least 2000 point clouds.

[0036] Optionally, in the visual method as described above, in step III, the comparison is achieved through the following steps:

[0037] Differentiate the empty conveyor belt point cloud model without coal from the coal quantity point cloud model to obtain a point cloud image of the coal quantity; judge the coal quantity based on the point cloud image of the coal quantity.

[0038] To achieve the foregoing objective, a second aspect of the present invention provides a computer program, wherein the computer program includes sequentially executing the steps of the visual method as described in any one of the foregoing first aspects.

[0039] To achieve the foregoing objective, a third aspect of the present invention provides a visual system for measuring the coal quantity on a conveyor belt of a belt conveyor, wherein the visual system includes:

[0040] An upper 3D binocular point cloud camera that captures a first 3D point cloud image above the conveyor belt;

[0041] A lower 3D binocular point cloud camera that captures a second 3D point cloud image below the conveyor belt;

[0042] A surface light source that provides a uniform illumination environment for the upper 3D binocular point cloud camera and the lower 3D binocular point cloud camera, and the surface light source is an active light source that projects structured light;

[0043] A strong light baffle that is used to block the interference of strong light on the structured light; and

[0044] A computer processing and control module that measures the coal quantity on the conveyor belt of the belt conveyor according to the visual method as described in any one of the foregoing first aspects and sends conveyor belt speed control information based on the measured coal quantity passing through.

[0045] To achieve the foregoing objective, a fourth aspect of the present invention provides a conveyor belt speed regulation system for a belt conveyor, wherein the speed regulation system includes the visual system as described in the foregoing third aspect and a conveyor belt speed controller, and the conveyor belt speed controller adjusts the speed of the conveyor belt according to the conveyor belt speed control information sent by the computer processing and control module.

[0046] The present invention can accurately obtain the amount of coal on the conveyor belt of a belt conveyor. The vision system for measuring the amount of coal on the conveyor belt of the present invention, through the cooperation of a 3D binocular point cloud camera, a surface light source, and a strong light baffle, is not affected by the intensity change of ambient light, and thus accurately acquires the three-dimensional point cloud image data of the amount of coal on the conveyor belt of the conveyor. According to the present invention, the amount of coal on the conveyor belt of the belt conveyor is measured and conveyor belt speed control information is sent based on the measured amount of coal, so as to increase the transportation speed of the conveyor belt when the amount of coal is large and slow down the transportation speed of the conveyor belt when the amount of coal is small, and the effect of green and energy-saving transportation of the conveyor can be achieved.

Description of the Drawings

[0048] Referring to the accompanying drawings, the disclosure of the present invention will become more apparent. It should be understood that these drawings are only for illustrative purposes and are not intended to limit the scope of protection of the present invention. In the figures:

[0049] Figure 1 is a schematic diagram of an embodiment of a conveyor belt speed control system for a belt conveyor according to the present invention, and an embodiment of a vision system for measuring the amount of coal on the conveyor belt of the belt conveyor according to the present invention is also shown in the figure;

[0050] Figure 2 is Figure 1 a cross-sectional schematic diagram of the vision system for measuring the amount of coal on the conveyor belt of the belt conveyor in, where point cloud acquisition is shown from the front; and

[0051] Figure 3 is a flowchart of an embodiment of a vision method for measuring the amount of coal on the conveyor belt of a belt conveyor according to the present invention.

[0052] Reference numerals: 100 - conveyor belt speed control system; 110 - vision system; 111 - upper 3D binocular point cloud camera; 112 - lower 3D binocular point cloud camera; 113 - computer processing and control module; 120 - conveyor belt speed controller; 130 - conveyor belt; 140 - coal; f1 - upper surface parametric equation; f2 - lower surface parametric equation.

Detailed Embodiments

[0054] Referring to the accompanying drawings, embodiments of a vision method for measuring the amount of coal on the conveyor belt of a belt conveyor, a computer program, a vision system for measuring the amount of coal on the conveyor belt of a belt conveyor, and a conveyor belt speed control system for a belt conveyor according to the present invention will be described by way of example below. However, all descriptions should not be used to form any limitation on the scope of protection of the present invention.

[0055] In addition, for any single technical feature described or implied in the embodiments mentioned herein, or any single technical feature shown or implied in the various drawings, the present invention still allows any combination or deletion to continue between these technical features (or their equivalents) without any technical obstacles, and thus it should be considered that these additional embodiments according to the present invention are also within the scope of the description herein.

[0056] It should also be noted that in the description of the present invention, the orientation or positional relationship indicated by terms such as "upper" and "lower" is based on the orientation or positional relationship in the upper and lower directions of the conveyor belt of the conveyor shown in the drawings. It is only for the convenience of describing the present disclosure and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus cannot be construed as a limitation to the present disclosure.

[0057] In addition, terms such as "first" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, features defined with "first", "second", etc. may explicitly or implicitly include at least one of such features.

[0058] Figure 1 FIG. 10 is a schematic diagram of an embodiment of a conveyor belt speed regulation system for a belt conveyor according to the present invention, and an embodiment of a vision system for measuring the coal quantity on the conveyor belt of the belt conveyor according to the present invention is also shown in the figure. In the coal production industry, when using the conveyor belt 130 to convey coal, the coal quantity can be detected by the vision system 110 and the conveyor belt speed can be controlled through the control end.

[0059] As can be seen from Figure 1 FIG. 10, the conveyor belt speed regulation system 100 may include a vision system 110 and a control end, and the control end may include a computer processing and control module 113. A conveyor belt speed controller 120 is also shown in the figure. The computer processing and control module 113 and the conveyor belt speed controller 120 may be integrated into the same module.

[0060] In the illustrated example, the vision system 110 includes an upper 3D binocular point cloud camera 111 and a lower 3D binocular point cloud camera 112. In alternative embodiments, other numbers of binocular point cloud cameras may be provided. The shooting ranges of the upper 3D binocular point cloud camera and the lower 3D binocular point cloud camera are shown in the figure. As can be seen from the figure, the shooting ranges of the upper 3D binocular point cloud camera and the lower 3D binocular point cloud camera correspond to each other and are located at the same position on the conveyor belt of the conveyor.

[0061] In this Figure 1In the embodiment, the control end includes a computer processing and control module 113, and the computer processing and control module 113 may further include a conveyor belt speed controller 120. The computer processing and control module 113 provides conveyor belt speed control information based on the measured coal quantity data on the conveyor belt, and the conveyor belt speed controller 120 adjusts the speed of the conveyor belt 130 based on this conveyor belt speed control information. Additionally, as shown in the figure, coal 140 is located on the conveyor belt 130 of the belt conveyor and is conveyed through this conveyor belt 130. When the coal quantity is large, the driving power can be increased, the conveying speed can be accelerated, and the work efficiency can be improved; when the coal quantity is small, the driving power can be reduced, the conveying speed can be slowed down, and the energy loss can be reduced.

[0062] The upper 3D binocular point cloud camera 111 of the vision system 110 can be directly above a specific position on the conveyor belt 130. This upper 3D binocular point cloud camera 111 is used to capture a three-dimensional point cloud image of a section of the conveyor belt 130 and / or the coal 140 along the forward direction of the conveyor belt 130 below it. A lower 3D binocular point cloud camera 112 is arranged directly below this specific position on the conveyor belt 130. The conveyor belt 130 is between the upper 3D binocular camera 111 and the lower 3D binocular camera 112. The specific position is a position where a binocular point cloud camera is installed at any arbitrarily selected point on the conveyor belt of the conveyor.

[0063] According to this setting, a certain range can be selected on the conveyor belt of a belt conveyor several kilometers long, and the coal quantity on the current belt conveyor can be continuously measured through the visual 3D point cloud volume measurement method, which can provide a decision for the system to adjust the belt speed of the conveyor belt and ensure the transportation efficiency of the belt conveyor.

[0064] In an alternative embodiment, this position can be selected near the entrance of the coal transportation on the conveyor belt. By taking pictures at this position, the binocular point cloud camera can know the exact volume of the coal 140 on the conveyor belt 130 from the start of the coal transportation, so as to more quickly and accurately adjust the conveying speed of the conveyor belt 130 according to this volume, so as to achieve the effect of green and energy-saving transportation of the conveyor. It is not excluded that in other embodiments, the binocular point cloud camera can be set at other positions. Setting this position near the entrance of the coal transportation on the conveyor belt is beneficial to accurately determining the conveyed coal quantity on the conveyor belt according to needs after measuring for a period of time.

[0065] It should be noted here that when the coal 140 is on the conveyor belt 130, its weight will cause the middle part of the conveyor belt 130 to sag, as Figure 2 shown.

[0066] Figure 2 For Figure 1Schematic cross-sectional view of a vision system for measuring the coal quantity on a conveyor belt of a belt conveyor, where point cloud acquisition is shown from the front. From Figure 2 it can be seen two exemplary arranged cameras of the upper 3D binocular point cloud camera 111 and the lower 3D binocular point cloud camera.

[0067] In this example, the upper 3D binocular point cloud camera 111 can capture the upper part of the coal 140. In order to capture the coal 140 in the concave part, a lower 3D binocular point cloud camera 112 is arranged at the position directly below the conveyor belt 130 corresponding to the upper 3D binocular point cloud camera 111 at the aforementioned specific position of the conveyor belt 130. The lower 3D binocular point cloud camera 112 is used to capture the three-dimensional point cloud image of the conveyor belt 130 and / or the coal 140 in the concave part at the same distance above it as captured by the upper 3D binocular point cloud camera 111. In the illustrated example, the upper 3D binocular point cloud camera and the lower 3D binocular point cloud camera respectively cover the entire width range of the conveyor belt, and all the coal and the downward concave conveyor belt are included in the shooting range in the width direction.

[0068] In this embodiment, the upper 3D binocular point cloud camera 111 and the lower 3D binocular point cloud camera 112 can adopt the Intel D435 Realsense depth camera, which has mature products, small volume, convenient flexibility, low resource consumption, and can calculate the depth map from a single-frame IR image with low power consumption. Moreover, the upper 3D binocular point cloud camera 111 and the lower 3D binocular point cloud camera 112 in this embodiment have high precision and resolution, and the effective frame rate can reach 60 FPS. Therefore, the combination of the upper 3D binocular point cloud camera 111 and the lower 3D binocular point cloud camera 112 can accurately collect and measure the point cloud data of the coal 140 on the conveyor belt 130. It should be noted here that the implementation of the present invention is not limited to using this kind of 3D binocular point cloud camera.

[0069] According to this embodiment, both the upper 3D binocular point cloud camera 111 and the lower 3D binocular point cloud camera 112 are depth cameras, which obtain image information through the left and right two cameras, calculate the parallax, and can measure the depth, that is, the distance, from the object to the camera. In order to be able to work in the case of insufficient light, a projection structured light can also be equipped for the upper 3D binocular point cloud camera 111 and the lower 3D binocular point cloud camera 112. Structured light measurement belongs to active optical measurement. Usually, a structured light of a certain shape is projected onto the object surface, and the three-dimensional information of the object is obtained by detecting the offset distance of the structured light.

[0070] Compared with the stereo vision composed of ordinary binoculars, which has complex external geometric calibration and feature matching operations, the method of structured light measurement is selected, that is, light with certain structural features is projected onto the object to be photographed to obtain a three-dimensional structure, with moderate cost and high precision.

[0071] In an alternative embodiment, the vision system 110 may further include an additional surface light source (not shown) and a strong light baffle (not shown). The surface light source is an active light source for projecting structured light for the binocular point cloud camera. This surface light source can operate in low-light conditions to provide a uniform and stable lighting environment for the binocular point cloud camera, so as to obtain sufficient point cloud data. At the position above the conveyor belt, the surface light source irradiates the coal 140, and by using the diffuse reflection of the irregular and rough surface of the coal 140, sufficient three-dimensional point cloud data of the upper part of the coal 140 can be obtained. At the position below the conveyor belt, the surface light source irradiates the bottom of the conveyor belt 130, and by using the diffuse reflection of the irregular and rough surface of the concave part of the conveyor belt 130, sufficient three-dimensional point cloud data of the lower part of the coal 140 can be obtained. Here, according to requirements, more than one surface light source can be provided above and below the conveyor belt. In an alternative embodiment, the data information of the thickness of the conveyor belt 130 can be loaded, and the three-dimensional point cloud data collected by the lower 3D binocular point cloud camera 112 can be removed of the data information of the thickness of the conveyor belt 130 to obtain a more accurate coal volume of the lower part of the coal 140.

[0072] The strong light baffle can be used to block the interference of strong light on the structured light, so as to avoid generating excessive noise points in the collected three-dimensional point cloud data, thereby making the shooting of the upper 3D binocular point cloud camera 111 and the lower 3D binocular point cloud camera 112 clearer, and the collected data more accurate, so as to measure more accurate coal volume data. This is more beneficial in the case where the mine light source is single, mostly from the direct irradiation of the miner's lamp, and the uneven light causes the brightness of some areas to be too high, interfering with the line of sight.

[0073] According to Figure 1 and Figure 2 In the embodiments of, the vision system 110 further includes a computer processing and control module 113. As shown by the arrow in Figure 1 , the upper 3D binocular point cloud camera 111 and the lower 3D binocular point cloud camera 112 transmit the three-dimensional point cloud data captured and collected above and below at specific positions of the conveyor belt 130 to the computer processing and control module 113, and the computer processing and control module 113 measures and analyzes to obtain the upper surface parameter equation f1 of the upper part of the coal 140 and the lower surface parameter equation f2 of the lower part of the coal 140, as shown in Figure 2 , and calculates the total coal volume of the coal 140 on the conveyor belt 130.

[0074] According to the total coal quantity, the conveyor belt speed controller 120 sends coal quantity control information through the serial port. After data conversion of the interface by the serial port service, the conveyor belt speed control information is sent to the variable-frequency AC motor through wireless communication to achieve intelligent control of the conveyor belt speed. The conveyor belt speed controller 120 adjusts the speed of the conveyor belt 130 according to this control information: when the measured coal quantity is large, the transportation speed of the conveyor belt 130 is increased to quickly transport the coal 140 to the ground and prevent congestion caused by the excessive residence time of the coal 140 on the conveyor belt 130; when the measured coal quantity is small, the transportation speed of the conveyor belt 130 is decreased to achieve an energy-saving effect, and slowing down the transportation speed can also extend the service life of the belt conveyor.

[0075] The power of the belt conveyor is mainly determined according to the peak production capacity of the coal mine combined with the corresponding redundancy factor, and it often runs at a constant high speed after startup. However, the production capacity of the mining face is not uniform, which often leads to the problem of "using a big horse to pull a small cart" with the belt conveyor running at a low load or even no load at high speed, resulting in a large amount of electrical energy waste. According to the control method in the foregoing embodiment, adjusting the speed according to the change of the coal quantity loaded on the belt conveyor can effectively alleviate the above problems, and the detection accuracy is stable.

[0076] Figure 3 It is a flowchart of an embodiment of a visual method for measuring the coal quantity on the conveyor belt of a belt conveyor according to the present invention. The visual method includes the following steps I - step VIII executed sequentially.

[0077] In step I, a first three-dimensional point cloud image above and a second three-dimensional point cloud image below at a specific position of the conveyor belt are respectively collected by an upper 3D binocular point cloud camera and a lower 3D binocular point cloud camera.

[0078] In this embodiment, an upper 3D binocular point cloud camera 111 and a lower 3D binocular point cloud camera 112 are adopted (see Figure 1 ), the upper 3D binocular point cloud camera 111 takes a first three-dimensional point cloud image above the conveyor belt 130 to collect the upper data of the conveyor belt 130; the lower 3D binocular point cloud camera 112 takes a second three-dimensional point cloud image below the conveyor belt 130 to collect the lower data of the conveyor belt 130. In an alternative embodiment, other numbers of binocular point cloud cameras can also be selected to make the measurement more accurate.

[0079] The specific position is not a fixed position. It can be a position where an upper 3D binocular point cloud camera and a lower 3D binocular point cloud camera are installed above any arbitrarily selected point on the conveyor belt of the conveyor. For example, but not limited to, in an alternative embodiment, this position can be selected at the entrance of the conveyor belt close to coal conveyance. By taking pictures at this position, the accurate volume of the coal 140 on the conveyor belt 130 can be known from the start of coal conveyance, so that the conveying speed of the conveyor belt 130 can be adjusted more quickly and accurately according to this volume, so as to achieve the effect of green and energy-saving conveyance of the conveyor. This position can also be selected in the middle section or the end section of the conveyor belt.

[0080] According to a specific embodiment, during the measurement process, a surface light source can be used to provide a uniform illumination environment for the upper 3D binocular point cloud camera 111 and the lower 3D binocular point cloud camera 112. The upper 3D binocular point cloud camera 111 and the lower 3D binocular point cloud camera 112 utilize the diffuse reflection of the surface of the coal and the conveyor belt 130 through the surface light source to obtain the three-dimensional point cloud data of the coal. The surface light source can be an active light source that projects structured light.

[0081] In step II, a method of random downsampling is adopted to simplify the number of point clouds in the first three-dimensional point cloud image and the second three-dimensional point cloud image.

[0082] It should be noted that the scale of the point cloud source file is very large. When the resolution of a binocular point cloud camera is 640*480, three hundred thousand points can be generated. Therefore, the point cloud files collected by the upper 3D binocular point cloud camera 111 and the lower 3D binocular point cloud camera 112 will be very huge, and a large amount of storage space and computing resources will be occupied. In this example, a method of random downsampling is adopted to simplify the number of point clouds in the first three-dimensional point cloud image and the second three-dimensional point cloud image. For example, the point cloud model can be reduced to 1 / 4 of its original size, and at the same time, the error caused by noise can also be eliminated.

[0083] In step III, a coal volume point cloud model of the coal volume on the conveyor belt is established based on the simplified first three-dimensional point cloud image and the simplified second three-dimensional point cloud image. The coal volume point cloud model is compared with the point cloud model of the conveyor belt running idly without coal to determine whether there is coal on the current conveyor belt. If so, proceed to step IV; otherwise, return to step I.

[0084] According to this embodiment, the no-coal point cloud model of the conveyor belt 130 can be pre-loaded. The no-coal point cloud model can be obtained in advance and pre-stored in the computer processing and control module. When the measurement accuracy is satisfied, the model file can be appropriately downsampled to reduce its size. For example, in an alternative embodiment, the no-coal point cloud model can be an octree model. The octree model is a tree-like data structure that can represent the no-coal point cloud model of the conveyor belt 130 at a specific position with a unified and simple shape to simplify the point cloud file and improve the calculation efficiency.

[0085] In this step, the foreground and background of the 3D point cloud depth map are separated. First, the 3D point cloud map of the conveyor belt without coal is recorded, and then it is differentiated from the point cloud map of the conveyor belt with coal during the production process to obtain the coal quantity point cloud map. That is, the no-coal point cloud model of the conveyor belt is differentiated from the coal quantity point cloud model to obtain the point cloud image of the coal quantity, and then the coal quantity can be judged based on the point cloud image of the coal quantity. If there is no coal passing through, return to step I and continue to wait until coal passes through.

[0086] In step IV, a section of distance is selected along the forward direction of the conveyor belt to extract the simplified first point cloud above the conveyor belt, and the same section of distance is selected at the corresponding position along the forward direction of the conveyor belt to extract the simplified second point cloud below the conveyor belt. Here, the first point cloud is the point cloud formed by the coal quantity above the conveyor belt; the second point cloud is the point cloud formed below the conveyor belt.

[0087] In step V, the observation areas corresponding to the first point cloud and the second point cloud are divided into point cloud grids, and the number of divided squares is m. Then, random downsampling of the point cloud is performed for each square of the first point cloud and the second point cloud, and the first point cloud and the second point cloud after the random downsampling of the point cloud are stored. Among them, m can be 100 of 10*10 or 128 of 16*8. In other alternative embodiments, the specific value of m can be selected according to needs.

[0088] The current observation area is divided into point cloud grids, and the number of grids is 100 - 128, which are 10*10 or 16*8, etc. (suitable for different camera resolutions). Then, random downsampling of the point cloud is performed for each square of the first point cloud and the second point cloud to further reduce the number of point clouds to 1 / 4 of the original. At this time, the existing points are stored, and at the same time, it can be ensured that the number of point clouds is at least 2000. If it does not meet the requirement, stop downsampling and then perform surface fitting.

[0089] In step VI, surface fitting is performed on the first point cloud above the conveyor belt in each square to obtain the upper surface parameter equation f1, and surface fitting is performed on the second point cloud below the conveyor belt in each square to obtain the lower surface parameter equation f2.

[0090] In step VI, the process of fitting the surface parameter equation f1 on the surface can be:

[0091] According to Y = AB, where A is [X1, X2, ..., X n ] T , B is [b0, b1, b2, b3, b4], Y is ln(z i ), X i is [1,x i , y i ,x i ², y i ²], B is the coefficient to be solved, where x i ,y i , z i is the coordinate of the i-th point cloud in the current grid, solve ;

[0092] Then the simultaneous equations

[0093]

[0094]

[0095]

[0096]

[0097]

[0098] Solving for x0, y0, z0 is

[0099]

[0100]

[0101] ,

[0102] Select x0, y0, z0 of each square, and then record the x0, y0, z0 calculated for the m squares as a i The parameter equation f1 is stored as the parameter equation representing the coal surface recorded by the 3D binocular point cloud camera above;

[0103] In step VI, the process of surface fitting the lower surface parameter equation f2 is:

[0104] For the point cloud formed by the 3D binocular point cloud camera below, the optimal solution is x1, y1, z1. The x0, y0 obtained by the process of surface fitting the upper surface parameter equation f1 are selected, and the corresponding nearest point cloud x1, y1, z1 is selected. Among them, x1, y1 are selected as , and then denote the calculated x0, y0, and z0 of the m squares as b i and store it as the parametric equation f2 representing the coal surface recorded by the lower 3D binocular point cloud camera.

[0105] Among them, m can be 100 of 10*10 or 128 of 16*8.

[0106] In step VII, the upper surface parametric equation f1 is represented by m point cloud points a1, a2, …, a m The lower surface parametric equation f2 is represented by m point cloud points b1, b2, …, b m Among them, the m point cloud points are spatial coordinates, and then use the horizontal positions of a1, a2, …, a m to correspond one by one with the horizontal positions of b1, b2, …, b m Taking the m point cloud points as spatial coordinates, divide the observation area of the coal volume point cloud model into m uniform squares, calculate the volume through the area ds of each square and the upper surface parametric equation f1 and the lower surface parametric equation f2 at the corresponding square, and sum them to obtain the passing volume of the coal on the conveyor belt at present . Among them, m can be 100 of 10*10 or 128 of 16*8.

[0107] Exemplarily, the upper surface parametric equation f1 can be represented by 100 point cloud points, which are a1, a2, …, a 100 , and these 100 point cloud points are spatial coordinates; and the lower surface parametric equation f2 can be represented by 100 point cloud points, which are b1, b2, …, b 100 , and these 100 point cloud points are spatial coordinates, where a1, a2, …, a 100 correspond one by one with b1, b2, …, b 100 on the longitudinal coordinate axis Z (as shown in Figure 2 ), XY is the horizontal coordinate axis, and Z is the longitudinal coordinate axis).

[0108] In step VIII, repeat steps I to VII, accumulate the passing volume each time to obtain the passing coal volume. What is obtained in step VII is the passing volume of the coal at a certain moment. The passing volume measured at multiple successive moments can obtain the passing coal volume on the conveyor belt after a period of time. In step VIII, repeat steps I to VII, accumulate the passing volume each time to obtain the passing coal volume. Each time steps I to VII are executed, the passing volume at a certain moment can be obtained; by executing steps I to VII multiple times, the passing volumes at multiple successive moments can be obtained, and accumulating them can obtain the passing coal volume.

[0109] The present invention further provides a computer program that can sequentially execute the steps of the foregoing visual method for measuring the coal quantity on the conveyor belt of a belt conveyor, as Figure 3 shown. Therefore, the computer program has each feature of the foregoing visual method and thus also has its corresponding advantages.

[0110] The coal production industry has always been a dangerous industry with a relatively high accident rate. The present invention provides an intelligent control system and method to replace manual observation and measurement, reduce the operating costs of enterprises, and adapt to the development of production from extensive to conservation-oriented. As an important part of coal production, compared with the conveyor system without load adjustment and running at full speed for a long time, the present invention can know the current coal production volume by visually detecting the coal quantity on a certain area of the conveyor belt, enabling it to achieve energy-saving operation, green transportation, increased production and efficiency. For example, it can save about 4.5 million to 6 million degrees of electricity, and save about 2 million yuan in conveyor belt maintenance costs, with huge economic benefits.

[0111] As described above, the present invention provides a visual method for measuring the coal quantity on the conveyor belt of a belt conveyor. Through this method, the coal quantity on the conveyor belt of the belt conveyor can be accurately obtained. The present invention further provides a computer program that can sequentially execute the steps of the foregoing visual method. The present invention also provides a visual system for measuring the coal quantity on the conveyor belt of a belt conveyor and a conveyor belt speed regulation system for a belt conveyor. These systems, through the cooperation of a binocular point cloud camera with a surface light source and a strong light baffle, are not affected by the intensity change of ambient light, so as to accurately collect the three-dimensional point cloud image data of the coal quantity on the conveyor belt of the conveyor. The computer processing and control module of the measurement system measures the coal quantity on the conveyor belt of the belt conveyor according to the foregoing visual method and sends conveyor belt speed control information based on the measured coal quantity, so as to increase the transportation speed of the conveyor belt when the coal quantity is large and slow down the transportation speed of the conveyor belt when the coal quantity is small, so as to achieve the effect of green and energy-saving transportation of the conveyor.

[0112] The technical scope of the present invention is not limited to the content in the above specification. Those skilled in the art can make various deformations and modifications to the above embodiments without departing from the technical idea of the present invention, and these deformations and modifications should all fall within the scope of the present invention.

Claims

1. A visual method for measuring the coal quantity on the conveyor belt of a belt conveyor, characterized in that, The visual method comprises the following steps performed sequentially: Step I: using an upper 3D binocular point cloud camera and a lower 3D binocular point cloud camera to respectively collect a first three-dimensional point cloud image above and a second three-dimensional point cloud image below a specific position of the conveyor belt; Step II: Simplifying the number of point clouds of the first three-dimensional point cloud image and the second three-dimensional point cloud image by using a random downsampling method; Step III: establishing a coal quantity point cloud model on the conveyor belt based on the simplified first three-dimensional point cloud image and the simplified second three-dimensional point cloud image, comparing the coal quantity point cloud model with the conveyor belt idling point cloud model without coal, and judging whether there is coal on the conveyor belt at present, if yes, proceed to step IV, otherwise return to step I; Step IV: selecting a distance along the advancing direction of the conveyor belt to extract a simplified first point cloud above the conveyor belt, and selecting the same distance at a corresponding position along the advancing direction of the conveyor belt to extract a simplified second point cloud below the conveyor belt; Step V: dividing the observation area corresponding to the first point cloud and the second point cloud into point cloud grids, the number of squares obtained by the division is m, and then randomly downsampling the point clouds for each square of the first point cloud and the second point cloud, and storing the first point cloud and the second point cloud after the point cloud random downsampling; Step VI: performing surface fitting on the first point cloud above the conveyor belt in each square to obtain an upper surface parameter equation f1, and performing surface fitting on the second point cloud below the conveyor belt in each square to obtain a lower surface parameter equation f2; Step VII: The upper surface parametric equation f1 is represented by m point cloud points a1, a2, …, a m The lower surface parametric equation f2 is represented by m point cloud points b1, b2, …, b m where the m point cloud points are spatial coordinates, and then the horizontal positions of b1, b2, …, b m are put into one-to-one correspondence with the horizontal positions of a1, a2, …, a m Taking the m point cloud points as spatial coordinates, the observation area of the coal quantity point cloud model is divided into m uniform grids. The volume is calculated and summed through the area ds of each grid and the upper surface parametric equation f1 and the lower surface parametric equation f2 at the corresponding grid, and the passing volume of the coal on the conveyor belt at present is obtained ; as well as Step VIII: Repeat steps I to VII, and add up the passing volume each time to obtain the passing amount of coal.

2. The visual method according to claim 1, wherein, In step VI, the process of surface fitting the upper surface parameter equation f1 is: According to Y = AB, where A is [X1, X2, ……, X n T , B is [b0, b1, b2, b3, b4], Y is ln(z i ), X i is [1, x i , y i , x i ², y i ²], B is the coefficient to be solved, where x i , y i , z i are the coordinates of the i-th point cloud in the current grid, solve ;​ Then the simultaneous equations Solving for x0, y0, z0 is ， Select x0, y0, and z0 for each grid, and then denote the x0, y0, and z0 calculated from the m grids as a i Store it as the parametric equation f1 of the upper surface representing the upper surface of the coal surface recorded by the upper 3D binocular point cloud camera; In step VI, the process of surface fitting the lower surface parameter equation f2 is: For the point cloud formed by the 3D binocular point cloud camera below, the optimal solution is x1, y1, z1. The x0, y0 obtained by the process of surface fitting the upper surface parameter equation f1 are selected, and the corresponding nearest point cloud x1, y1, z1 is selected. Among them, x1, y1 are selected as , and then denote the calculated x0, y0, and z0 of the m squares as b i and store it as the parametric equation f2 of the lower surface representing the lower surface of the coal surface recorded by the 3D binocular point cloud camera below.

3. The visual method according to claim 1, characterized in that m is 100 for 10*10 or 128 for 16*8.

4. The visual method according to claim 1, wherein During the measurement process, a surface light source is used to provide a uniform lighting environment for the upper 3D binocular point cloud camera and the lower 3D binocular point cloud camera. The upper 3D binocular point cloud camera and the lower 3D binocular point cloud camera utilize the diffuse reflection of the coal and the surface of the conveyor belt through the surface light source to obtain three-dimensional point cloud data of the coal.

5. The visual method according to claim 4, wherein, The surface light source is an active light source that projects structured light.

6. The visual method according to claim 1, characterized in that, The simplification in step II reduces the number of point clouds of the first three-dimensional point cloud image and the second three-dimensional point cloud image to 1 / 4 of the original size, and the downsampling in step V reduces the number of point clouds to 1 / 4 of the original size but to a minimum of 2000 point clouds.

7. The visual method according to claim 1, wherein, In step III, the comparison is achieved by the following steps: Differentiate the no-coal point cloud model of the conveyor belt during idling from the coal quantity point cloud model to obtain a point cloud image of the coal quantity; judge the coal quantity based on the point cloud image of the coal quantity.

8. A computer program, characterized in that, The computer program includes sequentially executing the steps of the vision method according to any one of the preceding claims 1 to 7.

9. A vision system for measuring the coal quantity on the conveyor belt of a belt conveyor, characterized in that, The vision system includes: An upper 3D binocular point cloud camera that captures a first three-dimensional point cloud image above the conveyor belt; A lower 3D binocular point cloud camera that captures a second three-dimensional point cloud image below the conveyor belt; A surface light source that provides a uniform lighting environment for the upper 3D binocular point cloud camera and the lower 3D binocular point cloud camera, and the surface light source is an active light source that projects structured light; A strong light baffle that is used to block the interference of strong light on the structured light; and A computer processing and control module that measures the coal quantity on the conveyor belt of the belt conveyor according to the vision method according to any one of the preceding claims 1 to 7 and sends conveyor belt speed control information based on the measured coal quantity passing through.

10. A conveyor belt speed regulation system for a belt conveyor, characterized in that, The speed regulation system includes the vision system according to claim 9 and a conveyor belt speed controller, and the conveyor belt speed controller adjusts the speed of the conveyor belt according to the conveyor belt speed control information sent by the computer processing and control module.

Images