Real-time flow detection method and system for belt conveying bulk materials and terminal

Abstract

The invention provides a belt conveying bulk material flow real-time detection method and system and a terminal, and the method comprises the steps: collecting a multi-frame bulk material contour image in real time through a double-line laser and an industrial camera, and recognizing a first line laser section contour and a second line laser section contour in the image, calculating the cross section area of each bulk material cross section, and constructing bulk material contour cross section time sequence data; through a sliding window time sequence analysis technology, time sequence continuity constraint is added, dynamic fitting and smooth processing are carried out on bulk material contour section time sequence data, and the instantaneous flow rate and the accumulated flow rate of the bulk materials are calculated in real time; therefore, non-contact, real-time, continuous, stable and credible bulk material conveying on-line detection under the severe operation conditions of complex illumination, much dust and the like is achieved, and the technical problems that an existing bulk material flow measurement method is low in measurement result precision, poor in stability, high in cost, high in maintenance difficulty and the like are solved.

Description

Technical Field

[0001] This application relates to the field of bulk material flow measurement technology, and in particular to a method, system and terminal for real-time detection of bulk material flow in belt conveyors. Background Technology

[0002] Belt conveyors are the most commonly used core equipment in the field of bulk material transportation, and are widely used in the continuous transport of bulk materials in industries such as solid waste disposal, mining, metallurgy, power, and ports. Accurate measurement of bulk material flow rate is not only crucial for intelligent speed regulation and energy-saving control of belt conveyors, but also an important foundation for realizing remote monitoring and digital management of material transport and production lines.

[0003] Traditional methods for measuring the flow rate of bulk materials rely on weighing equipment such as belt scales and nuclear scales. This involves measuring the weight of the bulk material on the belt at a given moment and combining this weight with the belt speed to calculate the instantaneous flow rate. While this method offers high accuracy, it is susceptible to factors such as the sensitivity of the weighing sensor, installation conditions, belt misalignment, vibration, and shock, leading to unstable measurement results and high maintenance costs.

[0004] To overcome the limitations of weighing measurement methods, researchers have proposed non-contact methods for measuring the flow rate of bulk materials based on video image analysis or laser scanning. These methods utilize single-line laser or structured light scanning technology to scan the bulk material on a conveyor belt, and one or more cameras capture surface profile images of the material to estimate the cross-sectional area and flow rate. However, due to the harsh environment of industrial sites, such as high dust levels, large variations in lighting, and irregular surfaces of the bulk material, the surface profile images captured by the cameras are prone to problems such as image blurring, light band interference, and incomplete cross-sectional extraction, leading to insufficient stability and robustness of the measurement results.

[0005] In recent years, with the development of sensing and computer vision technologies, researchers have begun to explore combining laser contour sensors with binocular vision technology to reconstruct the contours of bulk materials on conveyor belts. By comparing the contours with those of an empty conveyor belt, they can obtain the cross-sectional data of the bulk materials and then estimate the flow rate by combining this with the belt speed. For example, existing technologies disclose a real-time material flow detection scheme based on line light emission and parallel light illumination combined with dual cameras, which can adapt to some low-light environments and improve the accuracy of cross-section extraction. However, such schemes still suffer from problems such as complex hardware installation, high dependence on the empty belt calibration, and insufficient robustness of contour fitting algorithms to abnormal material shapes. Furthermore, they often require high-performance cameras and complex binocular matching calculations, increasing system cost and maintenance difficulty.

[0006] Therefore, with the development of industrial vision and image processing technology, non-contact bulk material flow measurement methods based on laser and vision fusion have gradually become a research hotspot. However, in industrial environments with complex lighting, variable material shapes, and severe on-site interference, how to extract clear and stable laser line features to ensure the accuracy of bulk material flow measurement by belt conveyor, and reduce costs and maintenance difficulties, remains a technical challenge that the industry urgently needs to overcome. Summary of the Invention

[0007] In view of the shortcomings of the prior art described above, the purpose of this application is to provide a method, system and terminal for real-time detection of the flow rate of bulk materials conveyed by belt conveyors, so as to solve the technical problems of low accuracy, poor stability, high cost and high maintenance difficulty of existing bulk material flow rate measurement methods.

[0008] To achieve the aforementioned and other related objectives, a first aspect of this application provides a method for real-time detection of the flow rate of bulk materials conveyed by a belt conveyor. The method includes: acquiring multiple frames of bulk material contour images in real-time using an industrial camera at a preset frame rate; extracting line laser features from each frame of the bulk material contour image to identify the first and second line laser cross-sectional contours formed by the projection of a first and second line laser beam generated by a dual-line laser onto the surface of the bulk material in each image; determining multiple bulk material cross-sections based on the first and second line laser cross-sectional contours in each frame of the bulk material contour image, calculating the cross-sectional area of ​​each cross-section, and constructing a time series data of the bulk material contour cross-sections; constructing multiple preset time-period models for the change in the height of the bulk material cross-sections based on the time series data, and correcting the cross-sectional area of ​​each bulk material cross-section and the time series data of the bulk material contour cross-sections; and calculating the instantaneous flow rate and cumulative flow rate of the bulk material in real-time based on the corrected time series data of the bulk material contour cross-sections.

[0009] In some embodiments of the first aspect of this application, the method for extracting line laser features from the bulk material contour image includes: cropping the bulk material contour image to obtain a line laser region of interest (ROI) image; performing smoothing, denoising, contrast enhancement, and binarization processing on the line laser ROI image to obtain a line laser ROI binary image; further denoising and smoothing processing on the line laser ROI binary image to obtain a line laser ROI feature image; and performing edge detection on the line laser ROI feature image to identify the first line laser cross-sectional contour and the second line laser cross-sectional contour.

[0010] In some embodiments of the first aspect of this application, the calculation method for the cross-sectional area of ​​the bulk material cross section includes: arranging multiple parallel sampling lines along the width direction of the belt to perform multi-point sampling on the first line laser cross section profile and the second line laser cross section profile respectively; obtaining the pixel coordinates of the first sampling point and the second sampling point intersecting with the first line laser cross section profile and the second line laser cross section profile respectively on each sampling line, and calculating the pixel distance between the first sampling point and the second sampling point of each sampling line accordingly, so as to calculate the cross-sectional height of the bulk material cross section at each sampling position; performing outlier removal and smoothing processing on each cross-sectional height of the bulk material cross section, reconstructing the cross-sectional curve of the bulk material cross section, and using integral operation to calculate the cross-sectional area of ​​the bulk material cross section.

[0011] In some embodiments of the first aspect of this application, before detecting the flow rate of bulk materials conveyed by the belt conveyor, the method further includes: calibrating the installation parameters of the industrial camera and the dual-line laser; specifically, the method includes: calibrating the industrial camera using a calibration plate containing a fixed-space pattern array to collect the pixel distance of the fixed-space pattern at multiple camera placement heights, thereby constructing a pixel distance conversion model; vertically placing the industrial camera and the dual-line laser at the same height above the belt, and determining the internal parameters of the industrial camera and the placement angle and spacing of the dual-line laser based on the placement height, so that the first and second line laser beams generated by the dual-line laser intersect on the belt surface.

[0012] In some embodiments of the first aspect of this application, the calculation method for the cross-sectional height of the bulk material cross section at a sampling position includes: based on the pixel distance conversion model, calculating the actual distance between the first sampling point and the second sampling point at the sampling position according to the pixel distance between the first sampling point and the second sampling point of the bulk material cross section; based on similar triangles, calculating the cross-sectional height of the bulk material cross section at the sampling position according to the actual distance between the first sampling point and the second sampling point, the calculation formula is: ;in, The cross-section of the bulk material at the first The cross-sectional height at each sampling location; The cross-section of the bulk material at the first The actual distance between the first and second sampling points at each sampling location; The spacing between the two-line lasers; The height at which the dual-line laser is deployed.

[0013] In some embodiments of the first aspect of this application, the method of constructing the time series data of the profile cross section of bulk materials includes: obtaining a first corrected cross section height at multiple sampling point positions based on the cross section curve of the bulk material cross section obtained from each frame of the bulk material profile image, and constructing the bulk material profile cross section height data for each bulk material cross section; marking each bulk material profile cross section height data according to the sampling time, position index, and cross section area of ​​the corresponding bulk material cross section of each frame of the bulk material profile image; dynamically stitching together the bulk material cross section height data at each sampling time and performing time-domain smoothing processing to construct the time series data of the bulk material profile cross section.

[0014] In some embodiments of the first aspect of this application, the method of constructing multiple models of bulk material cross-sectional height variation over multiple preset time periods and correcting the cross-sectional area of ​​each bulk material cross-section and the time series data of the bulk material profile cross-section includes: using a sliding window to perform segmented aggregation processing on multiple consecutive bulk material cross-sectional height data; performing dynamic fitting and smoothing processing on multiple bulk material cross-sectional height data within each window, and fitting and modeling the cross-sectional height variation trend of the bulk material cross-section within the preset time period corresponding to each window to obtain multiple models of bulk material cross-sectional height variation over multiple preset time periods; based on each model of bulk material cross-sectional height variation, obtaining the corrected cross-sectional curve of the bulk material cross-section at each sampling time and the second corrected cross-sectional height at multiple sampling point positions, and using integral calculation to calculate the corrected cross-sectional area of ​​each bulk material cross-section.

[0015] In some embodiments of the first aspect of this application, the instantaneous flow rate and cumulative flow rate of the bulk material are calculated in the following ways: ; ;in, for The first unit of time The corrected cross-sectional area of ​​the corresponding bulk material cross-section in the bulk material contour image acquired at each sampling time; for The number of samples of the outline image of bulk material per unit time; for The speed of the belt per unit time. , , They are respectively Time and Location index of the constantly acquired bulk material contour image; for The instantaneous flow rate of bulk materials per unit time; for to The cumulative flow of bulk materials within a given time period.

[0016] To achieve the aforementioned and other related objectives, a second aspect of this application provides a real-time detection system for the flow rate of bulk materials conveyed by a belt conveyor. The system includes: an image acquisition module comprising an industrial camera and a dual-line laser, used to acquire multiple frames of bulk material contour images in real time at a preset frame rate using the industrial camera, and to extract line laser features from each frame of the bulk material contour image, identifying the first and second line laser cross-sectional contours formed by the first and second line laser beams generated by the dual-line laser projected onto the surface of the bulk material in each bulk material contour image; and a cross-sectional data construction module connected to the image acquisition module, used to construct cross-sectional data based on each frame of the bulk material... The first and second laser cross-sectional profiles in the material profile image are used to determine multiple bulk material cross-sections, and the cross-sectional area of ​​each bulk material cross-section is calculated to construct time series data of the bulk material profile cross-sections. A cross-section data correction module, connected to the cross-section data construction module, is used to construct multiple preset time period bulk material cross-section height change models based on the bulk material profile cross-section time series data, and correct the cross-sectional area of ​​each bulk material cross-section and the bulk material profile cross-section time series data. A flow calculation module, connected to the cross-section data correction module, is used to calculate the instantaneous flow rate and cumulative flow rate of the bulk material in real time based on the corrected bulk material profile cross-section time series data.

[0017] To achieve the aforementioned and other related objectives, a third aspect of this application provides a real-time detection terminal for the flow rate of bulk materials conveyed by a belt conveyor. The real-time detection terminal for the flow rate of bulk materials conveyed by a belt conveyor includes: a processor and a memory; the memory is used to store a computer program; the processor is used to execute the computer program stored in the memory to cause the terminal to perform the real-time detection method for the flow rate of bulk materials conveyed by a belt conveyor as described in any of the above embodiments.

[0018] As described above, this application provides a method, system, and terminal for real-time detection of bulk material flow rate in belt conveyors. It uses a dual-line laser and an industrial camera to acquire multiple frames of bulk material contour images in real time, and identifies the first and second laser cross-sectional contours in the images to calculate the cross-sectional area of ​​each bulk material section, constructing a time series data of the bulk material contour cross-section. Through sliding window time series analysis technology, a time series continuity constraint is added to dynamically fit and smooth the bulk material contour cross-sectional time series data, allowing for real-time calculation of the instantaneous and cumulative flow rates of the bulk material. Therefore, this application has the following beneficial effects: it enables non-contact, real-time, continuous, stable, and reliable online detection of bulk material conveying under harsh operating conditions such as complex lighting and high dust levels, solving the technical problems of low measurement accuracy, poor stability, high cost, and high maintenance difficulty in existing bulk material flow rate measurement methods. Attached Figure Description

[0019] Figure 1 The diagram shown is a flowchart illustrating a method for real-time detection of the flow rate of bulk materials conveyed by belt conveyor, according to an embodiment of this application.

[0020] Figure 2 This diagram illustrates the correspondence between pixel distance and camera deployment height in one embodiment of this application.

[0021] Figure 3 The diagram shown illustrates the arrangement angle and spacing of the dual-line lasers in one embodiment of this application.

[0022] Figure 4 The diagram shown is a flowchart illustrating a method for calculating the cross-sectional area of ​​bulk materials in one embodiment of this application.

[0023] Figure 5 The diagram shown is a schematic representation of multiple sampling lines in one embodiment of this application.

[0024] Figure 6 The diagram shown illustrates the calculation of the cross-sectional height of a bulk material section in one embodiment of this application.

[0025] Figure 7 The diagram shown is a structural schematic of a real-time flow detection system for belt conveyed bulk materials according to an embodiment of this application.

[0026] Figure 8 The diagram shown is a structural schematic of a real-time flow detection terminal for belt conveyed bulk materials in one embodiment of this application. Detailed Implementation

[0027] The following specific examples illustrate the implementation of this application. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. This application can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this application. It should be noted that, unless otherwise specified, the following embodiments and features in the embodiments can be combined with each other.

[0028] In the embodiments of this application, terms such as "first" and "second" are used to distinguish identical or similar items with essentially the same function and effect. For example, the first sampling point and the second sampling point are only used to distinguish different sampling points and do not limit their order. Those skilled in the art will understand that terms such as "first" and "second" do not limit the quantity or execution order, and that terms such as "first" and "second" do not necessarily imply that they are different.

[0029] To address the problems mentioned in the background art, this application provides a method, system, and terminal for real-time detection of the flow rate of bulk materials conveyed by belt conveyors, aiming to solve the technical problems of low measurement accuracy, poor stability, high cost, and high maintenance difficulty in existing bulk material flow rate measurement methods.

[0030] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions in the embodiments of this application are further described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only for explaining this application and are not intended to limit this application.

[0031] like Figure 1 The diagram illustrates a flowchart of a method for real-time detection of the flow rate of bulk materials conveyed by a belt conveyor, according to an embodiment of this application. The method for real-time detection of the flow rate of bulk materials conveyed by a belt conveyor in this embodiment mainly includes the following steps.

[0032] Step S1: Acquire multiple frames of bulk material contour images in real time using an industrial camera at a preset frame rate, and extract line laser features from each frame of bulk material contour image to identify the first line laser cross-sectional contour and the second line laser cross-sectional contour formed by the first and second line laser beams generated by the dual-line lasers projected onto the surface of the bulk material in each bulk material contour image.

[0033] It should be understood that industrial cameras can convert captured light signals into ordered electrical signals through photoelectric conversion, which are then used as imaging devices to acquire contour images of bulk materials during belt conveyor operation. Preferably, an industrial camera with resolution and frame rate that meet the requirements of industrial field operations can be selected. Frame rate refers to the number of frames of the bulk material contour image that the industrial camera can acquire per second.

[0034] A dual-line laser is a device capable of simultaneously generating two line laser beams of different wavelengths or with a specific spatial structure. Preferably, a dual-line laser with stable wavelength and uniform beam pattern can be selected.

[0035] This application employs an industrial camera and a dual-line laser device, greatly simplifying the hardware structure and significantly reducing reliance on complex binocular devices and measurement systems. Furthermore, the dual-line laser can generate two beams of visible or infrared laser light. This results in lower cost and stronger contrast between visible and infrared light. When the first and second line laser beams generated by the dual-line laser are projected onto the surface of the bulk material, two clear light bands are formed to represent the contour of the bulk material's cross-section. This avoids problems such as image blurring and light band interference in the acquired bulk material contour image, reduces image processing difficulty, and ensures the accuracy and stability of the bulk material flow detection results based on the bulk material contour image.

[0036] In this embodiment, the first and second line laser beams have specific relative positions in space, such as being arranged at a certain spacing and angle. Each line laser beam includes multiple parallel laser lines that can cover the full width of the belt cross-section, thereby forming a first line laser cross-sectional profile and a second line laser cross-sectional profile that can represent the complete contour shape of the bulk material cross-section.

[0037] In one embodiment, before detecting the flow rate of bulk materials conveyed by the belt conveyor, the method further includes: calibrating the installation parameters of the industrial camera and the dual-line laser to ensure that the acquired outline image of the bulk materials has sufficient clarity and that there is a unique correspondence between the pixel coordinates of each pixel in the image and its physical size, so as to ensure the accuracy of the final bulk material flow rate detection result.

[0038] In this embodiment, the specific installation parameters include: the internal parameters and external attitude parameters (such as the installation height) of the industrial camera, as well as the installation spacing, installation angle, and installation height of the dual-line laser.

[0039] The specific calibration method includes the following steps.

[0040] ① The industrial camera is calibrated using a calibration plate containing a fixed-spacing pattern array to collect the pixel distance of the fixed-spacing pattern when multiple cameras are deployed at different heights, thereby constructing a pixel distance conversion model.

[0041] Specifically, pixel distances of fixed-interval patterns at multiple camera deployment heights are collected to construct multiple sets of correspondences between camera deployment heights and pixel distances, obtaining multiple data pairs. Each data pair includes the camera deployment height and the corresponding pixel distance of the fixed-interval pattern. A coordinate system is constructed with camera deployment height as the abscissa and pixel distance as the ordinate, and a scatter plot of each data pair is drawn in the coordinate system. Curve fitting is performed on each discrete point, and the fitted curve is drawn in the same coordinate system to construct the pixel distance conversion model, such as... Figure 2 As shown.

[0042] like Figure 2 As shown, the actual physical distance of the fixed spacing pattern is 30mm. When the camera is set up at a height of 0.6m, the corresponding pixel distance is 115, that is, the length of 115 pixels in the image captured by the industrial camera represents the actual physical distance of 30mm; when the camera is set up at a height of 0.4m, the corresponding pixel distance is 52, that is, the length of 52 pixels in the image captured by the industrial camera represents the actual physical distance of 30mm.

[0043] The pixel distance conversion model is used to convert the pixel positions in the bulk material contour image into actual physical distances, so as to calculate the cross-sectional area of ​​the bulk material and the bulk material flow rate. To improve the conversion accuracy, a large number of data pairs can be constructed, and various curve fitting methods, such as polynomial fitting and spline fitting, can be used to ensure the accuracy of the pixel distance conversion model. However, it should be noted that the number of data pairs and the curve fitting method can be selected by the user as needed, and this application does not specifically limit them.

[0044] ② The industrial camera and the dual-line laser are vertically positioned at the same height above the belt. Based on this positioning height, the internal parameters of the industrial camera and the positioning angle and spacing of the dual-line laser are determined so that the first and second line laser beams generated by the dual-line laser intersect on the surface of the belt.

[0045] Specifically, based on the belt width, the height range of the bulk material to be conveyed, and the installation environment of the industrial site, the installation height of the industrial camera and the installation height, angle, and spacing of the dual-line laser are reasonably determined. This simplifies the calibration process and avoids dust from the industrial site obstructing the industrial camera lens. It ensures that the first and second line laser beams always cover the full width of the belt cross-section, thus ensuring coverage of the effective cross-sectional area of ​​the bulk material. Simultaneously, it ensures that the first and second line laser beams intersect on the belt surface. Figure 3 As shown, this is to facilitate the subsequent calculation of the cross-sectional height of the bulk material based on the principle of similar triangles, and then to calculate the cross-sectional area of ​​the bulk material.

[0046] After calibrating the installation parameters of the industrial camera and the dual-line laser, the industrial camera then acquires multiple frames of bulk material contour images in real time at a preset frame rate. Line laser features are extracted from each frame of bulk material contour image to identify the first and second line laser cross-sectional contours in the image.

[0047] In one embodiment, the method for extracting line laser features from the outline image of the bulk material includes the following steps.

[0048] ① The outline image of the bulk material is cropped to obtain a line laser region of interest image.

[0049] The region of interest image of the line laser retains two light bands formed by the first and second line laser beams. In this embodiment, image cropping can effectively reduce interference from irrelevant backgrounds.

[0050] ② The region of interest image of the line laser is subjected to smoothing, noise reduction, contrast enhancement and binarization processing to obtain a binary image of the region of interest of the line laser.

[0051] In one specific embodiment, a median filter is used to smooth and denoise the region of interest image of the line laser, so as to effectively suppress high-frequency noise such as dust on the surface of loose materials and scattered light spots, and better protect the image edges and retain the high-contrast details of the light band area.

[0052] After denoising the region of interest image of the line laser, the image is further enhanced using the CLAHE (Contrast Limited Adaptive Histogram Equalization) method. This adaptively improves the local contrast of the image, ensuring that the light band region still has sufficient grayscale differences on the surface of loose materials of different colors and roughness, while effectively suppressing noise.

[0053] After denoising and enhancing the region of interest image of the line laser, Otsu's Method is used to adaptively determine the optimal segmentation threshold of the image by maximizing the inter-class variance or minimizing the intra-class variance, dividing the image into foreground and background classes. The image is then further binarized to obtain a binary image of the region of interest of the line laser, thus achieving background suppression.

[0054] ③ Further denoising and smoothing are performed on the binary image of the region of interest of the line laser to obtain the feature image of the region of interest of the line laser.

[0055] In one specific embodiment, morphological operators such as opening and closing operations are used to remove small isolated noise points in the binary image of the region of interest of the line laser, smooth the contour of the light band region, and interpolate and connect locally broken light band regions to form a continuous light band region with clear edges.

[0056] ④ Perform edge detection on the feature image of the region of interest of the line laser to identify the cross-sectional contours of the first and second line lasers.

[0057] The core objective of edge detection is to identify regions in the feature image of the region of interest of the line laser where the brightness or color changes drastically, namely the two light band regions formed by the first and second line laser beams projected onto the surface of the bulk material, thereby identifying the cross-sectional contours of the first and second line laser beams.

[0058] In one specific embodiment, the Sobel operator or the Canny operator can be used to highlight potential edges in the image, and then the real edge points can be detected and located by setting a threshold or other methods, thereby obtaining the first line laser cross-sectional profile and the second line laser cross-sectional profile.

[0059] In this embodiment, considering the characteristics of industrial sites with high dust levels and large variations in lighting, this application adopts a phased computer vision processing flow. By sequentially performing steps such as image cropping, median filtering for noise reduction, adaptive contrast enhancement, background suppression, and smoothing of light band areas on the outline image of the bulk material, multi-level optimization processing of the image can be achieved. This ensures the recognizability of the first and second line laser cross-sectional outlines in complex on-site operating environments, improves the cross-sectional extraction accuracy, and thus improves the calculation accuracy of the subsequent bulk material cross-sectional area and bulk material flow rate.

[0060] Step S2: Based on the first line laser cross-section profile and the second line laser cross-section profile in each frame of bulk material profile image, determine multiple bulk material cross-sections, calculate the cross-sectional area of ​​each bulk material cross-section, and construct time series data of bulk material profile cross-sections.

[0061] The dual-line laser is vertically positioned above the belt, projecting a first and a second line laser beam onto the belt. When the belt is unloaded, the two line laser beams form two parallel light bands on the belt, perpendicular to the belt's running direction and along its width. When bulk material flows across the belt, the shape of the light bands formed by the two line laser beams changes, forming a first and a second line laser cross-sectional profile that represent the complete outline of the bulk material's cross-section. Figure 4 As shown.

[0062] Based on the first and second laser cross-sectional profiles in each frame of the bulk material contour image, multiple corresponding bulk material cross-sections are determined. It should be noted that... Figure 3 , Figure 5 as well as Figure 6 The schematic diagram only magnifies the distance between the two line laser beams to illustrate the technical principle. In reality, the angle between the first and second line laser beams is extremely small, and the distance between them and the surface of the bulk material is also extremely small. Therefore, by using the two line laser beams to cut through the bulk material, an extremely thin three-dimensional model of the bulk material can be obtained, which can then be abstracted as a cross-section as the cross-section of the bulk material outline image in the corresponding frame.

[0063] After determining the cross-section of the bulk material in the bulk material contour image, the cross-sectional area of ​​the bulk material cross-section is calculated. This area is then used to construct a time series data of the bulk material contour cross-section, based on the cross-sectional areas of multiple bulk material cross-sections obtained from each frame of the bulk material contour image. The time series data of the bulk material contour cross-section includes: the sampling time of each frame of the bulk material contour image, and the corresponding position index, cross-sectional area, and height data of the bulk material contour cross-section.

[0064] In one embodiment, such as Figure 4 As shown, the calculation method for the cross-sectional area of ​​the bulk material includes steps S21 to S23.

[0065] Step S21: Arrange multiple parallel sampling lines along the width of the belt to perform multi-point sampling on the first line laser cross-sectional profile and the second line laser cross-sectional profile in the bulk material contour image.

[0066] Each sampling line runs along the length of the belt, i.e., along the belt's direction of travel. At this point, each sampling line is perpendicular to the light band formed by the first and second beam lasers, i.e., perpendicular to the cross-sectional profiles of the first and second beam lasers. For example... Figure 5 As shown, each sampling line intersects the first line laser cross-sectional profile and the second line laser cross-sectional profile at the first sampling point and the second sampling point, respectively.

[0067] In this embodiment, by setting multiple sampling lines, the effective area between the two line laser beams is sampled at multiple points, thereby providing sufficient discrete data support for subsequent fitting of the cross-sectional curve based on the cross-sectional height and integral calculation of the cross-sectional area.

[0068] In one embodiment, the sampling positions of each sampling line can be flexibly arranged according to the belt width, bulk material accumulation characteristics and detection accuracy requirements. Typically, multiple sampling lines are arranged at equal intervals.

[0069] Preferably, if the bulk material has a complex accumulation pattern or drastic local height changes, the number of sampling points can be dynamically adjusted as needed, and the sampling lines in the local areas with drastic height changes can be adaptively densified to obtain a higher resolution of the bulk material cross-sectional profile, improve the accuracy of cross-sectional curve fitting, and enhance the accuracy of bulk material flow detection.

[0070] Step S22: Obtain the pixel coordinates of the first sampling point and the second sampling point that intersect the first laser cross-section profile and the second laser cross-section profile of each sampling line, and calculate the pixel distance between the first sampling point and the second sampling point of each sampling line to calculate the cross-sectional height of the bulk material at each sampling position.

[0071] After obtaining the first and second line laser cross-sectional profiles, the pixel coordinates of the first and second sampling points of each sampling line intersecting the first and second line laser cross-sectional profiles are then detected. Based on this, the pixel distance between each first and second sampling point is calculated. Thus, through multi-point sampling, a set of pixel distance sequences distributed along the width of the bulk material cross-section is constructed, providing sufficient discrete data support for the subsequent construction of the bulk material profile cross-sectional time series data and the calculation of bulk material flow rate.

[0072] Based on the pixel distance between the first and second sampling points of each sampling line, the cross-sectional height of the bulk material at each sampling position is calculated. For example, the cross-sectional height of the bulk material at the first sampling position is calculated based on the pixel distance between the first and second sampling points of the first sampling line; the cross-sectional height of the bulk material at the second sampling position is calculated based on the pixel distance between the first and second sampling points of the second sampling line, and so on.

[0073] In one embodiment, the calculation of the cross-sectional height of the bulk material at a sampling location includes the following steps.

[0074] ①Based on the pixel distance conversion model, the actual distance between the first sampling point and the second sampling point at the sampling location is calculated according to the pixel distance between the first sampling point and the second sampling point of the bulk material cross section at the sampling location.

[0075] Wherein, the pixel distance conversion model is as follows: Figure 2 As shown, this is obtained by calibrating the industrial camera in advance.

[0076] ②Based on similar triangles, the cross-sectional height of the bulk material at the sampling position is calculated according to the actual distance between the first sampling point and the second sampling point.

[0077] like Figure 6 As shown, when the surface of the bulk material blocks the light bands generated by the first and second beam lasers, the shape of the light bands changes due to the height of the bulk material. Based on the two beam lasers' sampling of the bulk material, an extremely thin three-dimensional model of the bulk material can be obtained. Furthermore, a cross-section is taken from this three-dimensional model along the sampling line direction, as shown... Figure 6 As shown. It should be noted that, Figure 6 For the sake of demonstration purposes only, in reality, bulk materials are generally granular or powdery, such as coal, sand, grain, cement, sugar blocks, ore, and food. When transported by belt, they usually accumulate on the surface of the conveyor belt and do not form gaps with the belt surface. The estimated value of the cross-sectional area can be obtained simply by the change in cross-sectional height.

[0078] Since the line laser always propagates in a straight line, and it has been determined that the first and second line laser beams intersect on the belt surface when the belt is unloaded, then as follows: Figure 6 As shown, the actual distance between the two line laser beams detected on the pixel plane and the spatial height difference between the surface of the bulk material and the industrial camera satisfy a perspective relationship of similar triangles. Therefore, this application calculates the cross-sectional height of the bulk material based on the principle of similar triangles.

[0079] The calculation formula is:

[0080] ;Formula (1)

[0081] in, The cross-section of the bulk material at the first The cross-sectional height at each sampling location; The cross-section of the bulk material at the first The actual distance between the first and second sampling points at each sampling location; The spacing between the two-line lasers; The height at which the dual-line laser is deployed.

[0082] It should be noted that when calibrating the dual-line laser, if the first and second beam lasers do not strictly intersect the belt surface, the calculation method of the cross-sectional height of the bulk material can be adjusted by detecting the distance between the intersection point and the belt surface.

[0083] ;Formula (2)

[0084] ;Formula (3)

[0085] in, The cross-section of the bulk material at the first The cross-sectional height at each sampling location; The cross-section of the bulk material at the first The actual distance between the first and second sampling points at each sampling location; The spacing between the two-line lasers; The height at which the dual-line laser is deployed; The distance between the intersection point and the belt surface. When the intersection point of the first beam laser and the second beam laser is lower than the belt plane, the cross-sectional height of the bulk material section can be calculated using formula (2); when the intersection point of the first beam laser and the second beam laser is higher than the belt plane but lower than the surface of the bulk material, the cross-sectional height of the bulk material section can be calculated using formula (3).

[0086] Step S23: Remove outliers and smooth the height of each section of the bulk material cross-section, reconstruct the cross-sectional curve of the bulk material cross-section, and calculate the cross-sectional area of ​​the bulk material cross-section using integral calculation.

[0087] Due to interference factors such as noise, surface protrusions of bulk materials in local areas, or sampling jitter, the cross-sectional curve of the bulk material in a single frame of the bulk material contour image may have local anomalies or contour fluctuations. Therefore, after completing the multi-point calculation of the bulk material cross-section, this application prioritizes the removal of anomalies and smoothing of each cross-section height to improve the extraction accuracy of the cross-sectional curve and enhance the accuracy and robustness of subsequent calculations of cross-sectional area and bulk material flow rate.

[0088] In one specific embodiment, algorithms such as threshold elimination, median-based outlier detection, or local fitting residual analysis can be used to detect discrete points with abnormal cross-sectional heights in the bulk material cross-section and replace or delete them; methods such as moving average filtering, low-pass filtering, or polynomial curve fitting can be used to smooth the discrete points of each cross-sectional height.

[0089] In this embodiment, the purpose of this application is to effectively suppress local high-frequency noise and random jumps in the cross-sectional curve reconstructed based on discrete points of each cross-sectional height, effectively maintain the continuity and physical consistency of the overall shape of the cross-sectional curve, ensure the extraction accuracy of the cross-sectional curve, and thus lay a data foundation for constructing cross-frame time series analysis.

[0090] When performing outlier removal and smoothing on the cross-sectional heights of the bulk material, the cross-sectional heights are corrected. Based on the first corrected cross-sectional height of the corrected sampling point position, interpolation or polynomial fitting is performed on each discrete point to generate a continuous cross-sectional curve of the bulk material. Based on each first corrected cross-sectional height, integral operation is used to calculate the cross-sectional area of ​​the bulk material.

[0091] In one specific embodiment, the trapezoidal integral method or Simpson's integral method can be used to accumulate and superimpose the first corrected cross-sectional height at each sampling position to obtain the cross-sectional area of ​​the bulk material on the cross-section of the bulk material.

[0092] It should be understood that the principle of the trapezoidal rule is: to integrate the entire interval... (i.e., the width range of the cross-section of the bulk material) is divided into There are three equal-width intervals, with the width between each interval being [missing information]. In each small interval, the true curve is replaced by a straight line connecting the two endpoints of the function curve (i.e., the upper and lower bases of the trapezoid), and then the sum of the areas of all these small trapezoids is calculated as an approximation of the integral.

[0093] The principle of Simpson's method of integration is: [The text abruptly ends here, likely due to an incomplete sentence or a formatting error.] (i.e., the width range of the cross-section of the bulk material) is divided into ( There are an even number of equal-width intervals, with the width between each interval being [missing value]. In every two adjacent intervals (i.e., a width of...), On the interval ( ), the original function is approximated by a parabola passing through the two endpoints and the midpoint of the interval, and then the sum of the areas of all these small trapezoids is calculated as the integral approximation.

[0094] It should be noted that this application does not specifically limit the integration method used; users can choose according to their needs.

[0095] To ensure the temporal continuity of the bulk material cross-sectional information during the belt conveyor process, this application adopts the above steps S21~S23 to obtain the bulk material cross-section, multiple cross-sectional height discrete points, cross-sectional curves and cross-sectional areas of each bulk material cross-section corresponding to the multi-frame bulk material contour images captured by the industrial camera, and constructs the time series data of the bulk material contour cross-section based on this.

[0096] In one embodiment, the method for constructing the time series data of the bulk material profile cross-section includes the following steps.

[0097] ① Based on the cross-sectional curve of the bulk material cross-section obtained from the outline image of each frame of bulk material, obtain the first corrected cross-sectional height of multiple sampling points, and construct the bulk material outline cross-sectional height data for each bulk material cross-section.

[0098] The bulk material profile cross-sectional height data includes: the first corrected cross-sectional height at multiple sampling locations corresponding to the bulk material cross-section.

[0099] ② Mark the height data of each bulk material profile section based on the sampling time, position index, and cross-sectional area of ​​the corresponding bulk material section in each frame of the bulk material profile image.

[0100] Specifically, the industrial camera continuously acquires multiple frames of bulk material contour images at a preset frame rate. Based on the known frame rate, the sampling time of each frame of bulk material contour image can be determined, and this sampling time is marked as the sampling time of the height data of each bulk material contour section.

[0101] Furthermore, multiple position calibration points can be added to the belt so that when the industrial camera acquires the outline image of the bulk material, each position calibration point can be captured simultaneously to determine the position index of the bulk material cross-section obtained based on each bulk material outline image, and the position index is marked as the position index of the height data of each bulk material outline cross-section.

[0102] ③ The cross-sectional height data of the bulk material at each sampling time are dynamically sequenced and smoothed in the time domain to construct the time series data of the bulk material profile cross-section.

[0103] In this embodiment, this application smoothly connects multiple bulk material cross-sectional height data over time to form a continuous trend of bulk material cross-sectional height change. Through time-series smoothing processing, random fluctuations and noise in the data are suppressed, and data quality is enhanced. This provides an input basis for the subsequent construction of a bulk material cross-sectional height change model, and improves the accuracy, stability and robustness of bulk material flow calculation.

[0104] Step S3: Based on the time series data of the bulk material profile cross-section, construct multiple preset time period models of bulk material cross-sectional height change, and correct the cross-sectional area of ​​each bulk material cross-section and the time series data of the bulk material profile cross-section.

[0105] In one embodiment, a sliding window time series analysis technique is used to dynamically fit and smooth multiple bulk material cross-sectional height data in the time series data of the bulk material profile cross-section, thereby constructing multiple preset time period cross-sectional height change models of bulk materials, and correcting multiple cross-sectional heights and cross-sectional areas of each bulk material cross-section, further improving the accuracy and robustness of bulk material flow rate calculation. The specific method includes the following steps.

[0106] ① A sliding window is used to perform segmented aggregation of cross-sectional height data of multiple consecutive bulk materials.

[0107] Specifically, based on the belt's running speed and the industrial camera's frame rate, appropriate sliding window length and step size parameters are set, such as setting the sliding window length to be consistent with the industrial camera's frame rate.

[0108] ② Dynamically fit and smooth the cross-sectional height data of multiple bulk materials in each window, and fit and model the cross-sectional height change trend of the bulk materials in each window within a preset time period to obtain multiple cross-sectional height change models of bulk materials in corresponding preset time periods.

[0109] Specifically, methods such as polynomial fitting, B-spline curve fitting, or Kalman filtering can be used to dynamically fit and smooth the height data of multiple bulk material cross sections within each window, including smoothing out abnormal jumps caused by local fluctuations or residual noise, thereby simulating the trend of height change of bulk material cross sections within a preset time period.

[0110] ③ Based on the height change model of each bulk material section, obtain the corrected section curve of the bulk material section at each sampling time and the second corrected section height at multiple sampling points, and use integral calculation to calculate the corrected section area of ​​each bulk material section.

[0111] For example, the frame rate of the industrial camera is set to 25 fps, meaning the industrial camera acquires 25 frames of the bulk material contour image per second. Based on this frame rate, the window size of the sliding window is set to 25, meaning each window includes 25 bulk material contour cross-sectional samples. Each bulk material contour cross-sectional sample includes a first corrected cross-sectional height at multiple sampling positions corresponding to the bulk material cross-section. Using methods such as polynomial fitting, B-spline curve fitting, or Kalman filtering, the first corrected cross-sectional heights of the multiple bulk material cross-sections at the same sampling position within each window are smoothed to obtain the height change trend of the bulk material cross-section at that sampling position. This simulates the overall height change trend of the bulk material cross-section per second, obtaining a model of the bulk material cross-section height change per second. This model is then used to correct each first corrected cross-sectional height, resulting in multiple corresponding second corrected cross-sectional heights. Based on each second corrected cross-sectional height of each bulk material cross-section, the cross-sectional curve of the bulk material cross-section is reconstructed, and integral operations are used to recalculate the corrected cross-sectional area of ​​the bulk material cross-section.

[0112] In this embodiment, the core significance of using a sliding window lies in introducing local smoothing constraints in the time dimension. This ensures that the cross-sectional information of the bulk material is not only continuous in space but also maintains a consistent trend in time. This effectively suppresses the influence of outliers in a single frame of the bulk material contour image on the instantaneous flow rate calculation, preventing the accumulation of noise errors from causing deviations in the cumulative flow rate of the bulk material. This ensures the accuracy, stability, and anti-interference capability of the bulk material flow rate calculation. Through continuous scrolling updates of the sliding window, a smoothed, fitted model of the bulk material cross-sectional height change can be output in real time, ensuring accurate and real-time calculation of the instantaneous and cumulative flow rates of the bulk material. This enables non-contact, real-time, continuous, stable, and reliable online detection of bulk material conveying under harsh industrial conditions such as complex lighting and high dust levels, providing accurate flow rate references for subsequent automatic speed regulation and intelligent control of the production process.

[0113] Step S4: Calculate the instantaneous flow rate and cumulative flow rate of the bulk material in real time based on the corrected time series data of the bulk material profile cross section.

[0114] In one embodiment, the formula for calculating the instantaneous flow rate of the bulk material is:

[0115] ;Formula (4)

[0116] The formula for calculating the cumulative flow of bulk materials is:

[0117] ;Formula (5)

[0118] in, for The first unit of time The corrected cross-sectional area of ​​the corresponding bulk material cross-section in the bulk material contour image acquired at each sampling time; for The number of samples of the outline image of bulk material per unit time; for The speed of the belt per unit time. , , They are respectively Time and Location index of the constantly acquired bulk material contour image; for The instantaneous flow rate of bulk materials per unit time; for to The cumulative flow of bulk materials within a given time period.

[0119] This application uses a dual-line laser and an industrial camera to acquire multiple frames of bulk material contour images. Computer vision technology is used to denoise, enhance, and extract line laser features from each frame of the bulk material contour image. By sampling at multiple points along the cross-sectional direction of the bulk material and using similar triangles, the cross-sectional height and area of ​​each bulk material cross-section are accurately calculated. Furthermore, by employing sliding window temporal analysis technology and adding temporal continuity constraints, dynamic fitting and smoothing are performed on the multiple bulk material contour cross-sectional height data obtained from the multiple frames of bulk material contour images. The instantaneous and cumulative flow rates of the bulk material are calculated in real time. This enables non-contact, real-time, continuous, stable, and reliable online detection of bulk material conveying under harsh industrial conditions such as complex lighting and high dust levels, providing accurate flow rate references for automatic speed regulation and intelligent control of the production process.

[0120] like Figure 7 The diagram shown illustrates the structure of a real-time flow detection system 700 for belt conveying bulk materials according to an embodiment of this application. The real-time flow detection system 700 for belt conveying bulk materials in this embodiment mainly includes: an image acquisition module 701, a cross-sectional data construction module 702, a cross-sectional data correction module 703, and a flow calculation module 704, connected in sequence.

[0121] Specifically, the image acquisition module 701 includes an industrial camera and a dual-line laser, used to acquire multiple frames of bulk material contour images in real time at a preset frame rate through the industrial camera, and to extract line laser features from each frame of bulk material contour image, and to identify the first line laser cross-sectional contour and the second line laser cross-sectional contour formed by the first and second line laser beams generated by the dual-line laser beams projected onto the surface of the bulk material in each bulk material contour image.

[0122] The cross-section data construction module 702 is used to determine multiple bulk material cross sections based on the first line laser cross section profile and the second line laser cross section profile in each frame of bulk material profile image, and to calculate the cross-sectional area of ​​each bulk material cross section to construct bulk material profile cross section time series data.

[0123] The cross-sectional data correction module 703 is used to construct multiple preset time period cross-sectional height change models of bulk materials based on the time series data of the bulk material profile cross-section, and to correct the cross-sectional area of ​​each bulk material cross-section and the time series data of the bulk material profile cross-section.

[0124] The flow calculation module 704 is used to calculate the instantaneous flow rate and cumulative flow rate of the bulk material in real time based on the corrected time series data of the bulk material profile cross section.

[0125] It should be understood that the specific process of each module performing the aforementioned corresponding steps has been described in detail in the aforementioned method embodiments, and will not be repeated here for the sake of brevity.

[0126] It should also be understood that the module division in the embodiments of this application is illustrative and only represents a logical functional division; in actual implementation, there may be other division methods. Furthermore, the functional modules in the various embodiments of this application can be integrated into a single processor, exist as separate physical entities, or be integrated into a single module. The aforementioned integrated modules can be implemented in hardware or as software functional modules.

[0127] Figure 8 This is a schematic diagram of the structure of the real-time flow detection terminal 800 for belt conveying bulk materials provided in an embodiment of this application. Figure 8 As shown, the real-time flow monitoring terminal 800 for bulk materials conveyed by a belt conveyor includes: at least one processor 801, a memory 802, at least one network interface 803, and a user interface 805. The various components in the terminal are coupled together via a bus system 804. It is understood that the bus system 804 is used to realize communication between these components. In addition to a data bus, the bus system 804 also includes a power bus, a control bus, and a status signal bus. However, for clarity, in... Figure 8 The general will label all buses as bus systems.

[0128] The user interface 805 may include a monitor, keyboard, mouse, trackball, clicker, button, touchpad, or touch screen.

[0129] It is understood that memory 802 can be volatile memory or non-volatile memory, or both. Non-volatile memory can be read-only memory (ROM) or programmable read-only memory (PROM), which serves as an external cache. By way of example, but not limitation, many forms of RAM are available, such as static random access memory (SRAM) and synchronous static random access memory (SSRAM). The memories described in the embodiments of this application are intended to include, but are not limited to, these and any other suitable categories of memory.

[0130] In this embodiment, the memory 802 is used to store various types of data to support the operation of the real-time detection terminal 800 for bulk material flow rate of the belt conveyor. Examples of this data include any executable program that operates on the real-time detection terminal 800 for bulk material flow rate of the belt conveyor, such as operating system 8021 and application program 8022. Operating system 8021 includes various system programs, such as framework layer, core library layer, driver layer, etc., for implementing various basic services and handling hardware-based tasks. Application program 8022 may include various applications, such as media player, browser, etc., for implementing various application services. The implementation of the real-time detection method for bulk material flow rate of the belt conveyor provided in this embodiment can be included in application program 8022.

[0131] The real-time flow detection method for bulk materials conveyed by belt conveyors disclosed in the foregoing embodiments of this application can be applied to, or implemented by, processor 801. Processor 801 may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the aforementioned method can be completed by the integrated logic circuitry of the hardware in processor 801 or by instructions in software form. The aforementioned processor 801 may be a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Processor 801 can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. General-purpose processor 801 may be a microprocessor or any conventional processor, etc. The steps of the real-time flow detection method for bulk materials conveyed by belt conveyors provided in the embodiments of this application can be directly manifested as execution by a hardware decoding processor, or execution by a combination of hardware and software modules in the decoding processor. The software modules may be located in a storage medium, which is located in a memory. The processor reads the information in the memory and, in conjunction with its hardware, completes the steps of the aforementioned method.

[0132] In an exemplary embodiment, the real-time flow detection terminal 800 for belt conveying bulk materials can be used by one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), or complex programmable logic devices (CPLDs) to execute the aforementioned real-time flow detection method for belt conveying bulk materials.

[0133] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the real-time detection method for the flow rate of bulk materials conveyed by belt conveyors described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0134] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0135] In summary, this application provides a method, system, and terminal for real-time detection of bulk material flow rate in belt conveyors. It uses a dual-line laser and an industrial camera to acquire multiple frames of bulk material contour images in real time, and identifies the first and second laser cross-sectional contours in the images to calculate the cross-sectional area of ​​each bulk material section, constructing a time series data of the bulk material contour cross-section. Through sliding window time series analysis technology, and by adding time series continuity constraints, the time series data of the bulk material contour cross-section is dynamically fitted and smoothed to calculate the instantaneous and cumulative flow rates of the bulk material in real time. This achieves non-contact, real-time, continuous, stable, and reliable online detection of bulk material conveying under harsh operating conditions such as complex lighting and high dust levels, solving the technical problems of low measurement accuracy, poor stability, high cost, and high maintenance difficulty in existing bulk material flow rate measurement methods.

[0136] Therefore, this application effectively overcomes the various shortcomings of the prior art and has high industrial application value.

[0137] The foregoing embodiments are merely illustrative of the principles and effects of this application and are not intended to limit this application. Any person skilled in the art can modify or alter the foregoing embodiments without departing from the spirit and scope of this application. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in this application should still be covered by the claims of this application.

Claims

1. A method for real-time detection of the flow rate of bulk materials conveyed by belt conveyor, characterized in that, include: The industrial camera acquires multiple frames of bulk material contour images in real time at a preset frame rate, and extracts line laser features from each frame of bulk material contour image to identify the first line laser cross-sectional contour and the second line laser cross-sectional contour formed by the first and second line laser beams generated by the dual line lasers projected onto the surface of the bulk material in each bulk material contour image. Based on the first and second laser cross-sectional profiles in each frame of bulk material profile image, multiple bulk material cross-sections are determined, and the cross-sectional area of ​​each bulk material cross-section is calculated to construct time series data of bulk material profile cross-sections. Based on the time series data of the bulk material profile cross section, construct multiple preset time period models of bulk material cross section height variation, and correct the cross section area of ​​each bulk material cross section and the time series data of the bulk material profile cross section. Based on the corrected time series data of the bulk material profile section, the instantaneous flow rate and cumulative flow rate of the bulk material are calculated in real time.

2. The method for real-time detection of the flow rate of bulk materials conveyed by belt conveyor according to claim 1, characterized in that, The methods for line laser feature extraction of the outline image of the bulk material include: The outline image of the bulk material is cropped to obtain a line laser region of interest image; The region of interest image of the line laser is subjected to smoothing, noise reduction, contrast enhancement, and binarization to obtain a binary image of the region of interest of the line laser. Further denoising and smoothing processing is performed on the binary image of the region of interest of the line laser to obtain the feature image of the region of interest of the line laser. Edge detection is performed on the feature image of the region of interest of the line laser to identify the cross-sectional contours of the first and second line lasers.

3. The method for real-time detection of the flow rate of bulk materials conveyed by belt conveyor according to claim 1, characterized in that, The calculation method for the cross-sectional area of ​​the bulk material includes: Multiple parallel sampling lines are laid out along the width of the belt to perform multi-point sampling on the profile of the first line laser cross section and the profile of the second line laser cross section respectively; The pixel coordinates of the first sampling point and the second sampling point that intersect the first laser cross-section profile and the second laser cross-section profile of each sampling line are obtained, and the pixel distance between the first sampling point and the second sampling point of each sampling line is calculated accordingly, so as to calculate the cross-sectional height of the bulk material at each sampling position. The cross-sectional heights of the bulk material are subjected to outlier removal and smoothing, the cross-sectional curves of the bulk material are reconstructed, and the cross-sectional area of ​​the bulk material is calculated by integral operation.

4. The method for real-time detection of the flow rate of bulk materials conveyed by belt conveyor according to claim 3, characterized in that, Before detecting the flow rate of bulk materials conveyed by a belt conveyor, the following steps are also included: calibrating the installation parameters of the industrial camera and the dual-line laser; the specific methods include: The industrial camera is calibrated using a calibration plate containing a fixed-spacing pattern array to collect the pixel distance of the fixed-spacing pattern when multiple cameras are deployed at different heights, thereby constructing a pixel distance conversion model. The industrial camera and the dual-line laser are vertically positioned at the same height above the belt. Based on this positioning height, the internal parameters of the industrial camera and the positioning angle and spacing of the dual-line laser are determined so that the first and second line laser beams generated by the dual-line laser intersect on the belt surface.

5. The method for real-time detection of the flow rate of bulk materials conveyed by belt conveyor according to claim 4, characterized in that, The calculation method for the cross-sectional height of the bulk material at a sampling location includes: Based on the pixel distance conversion model, the actual distance between the first sampling point and the second sampling point at the sampling location is calculated according to the pixel distance between the first sampling point and the second sampling point of the bulk material cross section at the sampling location. Based on similar triangles, and according to the actual distance between the first sampling point and the second sampling point, the cross-sectional height of the bulk material at the sampling location is calculated using the following formula: ； in, The cross-section of the bulk material at the first The cross-sectional height at each sampling location; The cross-section of the bulk material at the first The actual distance between the first and second sampling points at each sampling location; The spacing between the two-line lasers; The height at which the dual-line laser is deployed.

6. The method for real-time detection of the flow rate of bulk materials conveyed by belt conveyor according to claim 3, characterized in that, Methods for constructing time series data of bulk material profile sections include: Based on the cross-sectional curve of the bulk material cross-section obtained from each frame of bulk material contour image, the first corrected cross-sectional height of multiple sampling point positions is obtained, and the bulk material contour cross-sectional height data of each bulk material cross-section is constructed. Based on the sampling time, position index, and cross-sectional area of ​​the corresponding bulk material cross-section of each frame of bulk material contour image, the height data of each bulk material contour cross-section is marked; The cross-sectional height data of the bulk material at each sampling time are dynamically sequenced and then smoothed in the time domain to construct the time series data of the bulk material profile cross-section.

7. The method for real-time detection of the flow rate of bulk materials conveyed by belt conveyor according to claim 6, characterized in that, The methods for constructing models of cross-sectional height changes of bulk materials over multiple preset time periods, and for correcting the cross-sectional area of ​​each bulk material section and the time series data of the bulk material profile section, include: A sliding window is used to perform segmented aggregation of cross-sectional height data for multiple consecutive bulk materials. Dynamic fitting and smoothing are performed on the cross-sectional height data of multiple bulk materials in each window, and the cross-sectional height change trend of the bulk materials in each window within a preset time period is fitted and modeled to obtain multiple cross-sectional height change models of bulk materials in the corresponding preset time periods. Based on the height variation model of each bulk material section, the corrected section curve of the bulk material section at each sampling time and the second corrected section height at multiple sampling points are obtained, and the corrected section area of ​​each bulk material section is calculated by integral operation.

8. The method for real-time detection of the flow rate of bulk materials conveyed by belt conveyor according to claim 7, characterized in that, Methods for calculating the instantaneous and cumulative flow rates of bulk materials include: ； ； in, for The first unit of time The corrected cross-sectional area of ​​the corresponding bulk material cross-section in the bulk material contour image acquired at each sampling time; for The number of samples of the outline image of bulk material per unit time; for The speed of the belt per unit time. , , They are respectively Time and Location index of the constantly acquired bulk material contour image; for The instantaneous flow rate of bulk materials per unit time; for to The cumulative flow of bulk materials within a given time period.

9. A real-time flow detection system for bulk materials conveyed by belt conveyors, characterized in that, include: The image acquisition module includes an industrial camera and a dual-line laser, which is used to acquire multiple frames of bulk material contour images in real time at a preset frame rate through the industrial camera, and extract line laser features from each frame of bulk material contour image to identify the first line laser cross-sectional contour and the second line laser cross-sectional contour formed by the first and second line laser beams generated by the dual-line laser beams projected onto the surface of the bulk material in each bulk material contour image. The cross-section data construction module, connected to the image acquisition module, is used to determine multiple cross-sections of bulk materials based on the first line laser cross-section profile and the second line laser cross-section profile in each frame of bulk material profile image, and to calculate the cross-sectional area of ​​each bulk material cross-section to construct time series data of bulk material profile cross-sections. The cross-section data correction module, connected to the cross-section data construction module, is used to construct multiple preset time period cross-section height change models of bulk materials based on the time series data of the bulk material profile cross-section, and correct the cross-sectional area of ​​each bulk material cross-section and the time series data of the bulk material profile cross-section. The flow calculation module, connected to the cross-sectional data correction module, is used to calculate the instantaneous flow rate and cumulative flow rate of the bulk material in real time based on the corrected time series data of the bulk material profile cross-section.

10. A real-time flow detection terminal for bulk materials conveyed by belt conveyors, characterized in that, include: Processor and memory; The memory is used to store computer programs; The processor is used to execute the computer program stored in the memory, so that the terminal performs the real-time detection method for the flow rate of bulk materials conveyed by belt conveyor as described in any one of claims 1 to 8.

Images