Method for accurately measuring outline of loading and unloading pile body of port bulk cargo storage yard

Abstract

The invention discloses a port bulk cargo storage yard loading and unloading stack contour accurate measurement method, and particularly relates to the technical field of contour measurement, and the method comprises the steps: S1, first-stage site reference coordinate system layout, S2, second-stage dynamic reference network layout, S3, laser scanning equipment deployment and calibration, S4, hierarchical scanning data collection, and S5, third-stage reference check and data correction. S6, filtering fusion and three-dimensional reconstruction of the point cloud data; and S7, actual measurement verification and result output, and outputting the result after actual measurement verification. According to the method, a three-level linkage checking mechanism of the site fixed datum, the temporary dynamic reference points and the pile body surface feature points is established, so that the accuracy of contour measurement is greatly improved, and the problem that three-level reference linkage checking is difficult to realize through the site datum, the temporary dynamic points and the pile body features is solved; and the accuracy of contour measurement is greatly reduced.

Description

Technical Field

[0001] This invention relates to the field of contour measurement technology, and more specifically, to a method for accurately measuring the contour of a bulk cargo yard loading and unloading structure. Background Technology

[0002] The precise measurement method of measuring the outline of the loading and unloading stack in the port bulk cargo yard using optical methods has the advantages of improving management efficiency, ensuring operational safety, optimizing resource allocation, and supporting decision-making in port operations. The specific analysis is as follows: real-time monitoring and dynamic management improve operational efficiency, full-scenario coverage and automated data collection eliminate the need for manual intervention and significantly shorten the measurement cycle.

[0003] In publicly available literature, patent publication number CN107314741A discloses a method for measuring cargo volume. This technology processes points in the point cloud data of loaded and empty vehicles separately using a method of constructing triangular planes to obtain the volume information of loaded and empty vehicles respectively. The invention is not limited by the resolution, sensitivity, lens distortion, installation position, or adverse effects of environmental factors of the front-end image acquisition equipment. The acquired data is direct measurement data, which has the advantages of high accuracy, real-time, objectivity, and vividness, and is worth promoting. However, this technology still has the following problems.

[0004] When conducting contour measurements on unloading bodies in port bulk cargo yards, the volume typically ranges from several thousand to tens of thousands of cubic meters. The surface is irregularly curved due to the fluidity of bulk cargo, and the shape changes in real time due to loading and unloading operations. In addition, a large amount of dust is generated during the operation, causing changes in the measurement reference points. It is difficult to achieve three-level reference linkage verification through site benchmarks, temporary dynamics, and pile characteristics, which significantly reduces the accuracy of contour measurements. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the prior art, the present invention provides a method for accurately measuring the contour of a bulk cargo yard loading and unloading structure, comprising the following specific steps: S1. Establishment of the primary site reference coordinate system: Multiple non-collinear fixed reference stakes are set up to form a primary site reference coordinate system; S2. Secondary dynamic reference network deployment: Magnetic temporary reference points are deployed on the surface of the stack to form a secondary dynamic reference network; S3. Deployment and calibration of laser scanning equipment: Deploy laser scanning equipment at multiple stations and calibrate it based on a primary reference. S4. Layered scanning data acquisition: Layered scanning acquisition of the heap outline and two levels of reference point data; S5. Three-level reference verification and data correction: Through three-level reference verification, data is corrected by site benchmark, temporary reference and feature point; S6. Point cloud data filtering, fusion and 3D reconstruction: Filtering and fusing point cloud data and reconstructing the 3D contour; S7. Actual Measurement Verification and Result Output: Output the results after actual measurement verification, and keep the measurement error within a reduced range throughout the process.

[0006] Preferably, in S1, a laser reflective target ball is embedded at the top of the fixed reference pile, and the positioning is calibrated using GPS / BeiDou dual-mode.

[0007] Preferably, the temporary reference point in S2 is a magnetically attached wear-resistant reference ball, which is arranged at different densities in the non-critical areas and critical contour areas of the pile body, and maintains a reasonable straight-line distance from the fixed reference pile.

[0008] Preferably, the laser scanning device in S3 is a three-dimensional laser scanner, with multiple devices deployed collaboratively. Each device covers a certain number of fixed benchmark piles and temporary reference points to ensure data synchronization between devices.

[0009] Preferably, in step S4, a layered scanning mode is adopted to control the overlap rate of the scanning areas of adjacent stations, and the dust removal and blowing device is activated when the dust concentration reaches a certain level.

[0010] Preferably, in step S5, a three-level reference verification standard is set to clarify the threshold for removing outliers and the control requirements for the proportion of outliers, so as to ensure the accuracy of data correction.

[0011] Preferably, in step S6, a Gaussian filtering algorithm is used to process the point cloud data, and the three-dimensional contour is reconstructed through triangular mesh modeling, taking into account both noise removal and detail preservation, so as to ensure data validity and contour restoration accuracy.

[0012] Preferably, in step S7, feature points in different regions are selected for actual measurement and verification, and a handheld rangefinder is used to assist in the verification to ensure that the volume calculation error meets the requirements.

[0013] Preferably, the stockpile is a bulk cargo such as coal or ore, which is suitable for different dust concentration conditions at the port and can achieve rapid and accurate measurement of a single stockpile.

[0014] Preferably, in step S5, the least squares method is used to fit the characteristic curve of the pile surface. Through feature point screening, model construction, error equation establishment, coefficient solution, and fitting verification and correction, the fitted curve is ensured to fit the measured feature points.

[0015] The technical effects and advantages of this invention are as follows: 1. This invention establishes a three-level linkage verification mechanism, consisting of a fixed site benchmark, temporary dynamic reference points, and feature points on the surface of the stack, thereby constructing a complete spatial coordinate verification system. This system enables real-time mutual verification and correction between multiple levels of references, effectively overcoming the inaccuracy of reference points caused by the mobility of bulk cargo, dust interference, and dynamic changes in operations. By using site benchmarks, temporary dynamic references, and stack features to achieve three-level linkage verification, the reliability and anti-interference capability of the overall measurement are significantly improved, and the accuracy of contour measurement is greatly enhanced.

[0016] 2. This invention adopts a layered scanning and multi-station collaborative data acquisition strategy, combined with efficient dust removal methods, to achieve blind-spot-free and high-density data acquisition of the pile body contour under complex working conditions. At the same time, through advanced point cloud processing algorithms, it performs intelligent filtering, precise stitching and surface reconstruction on massive scanning data, ensuring that the final three-dimensional model can reproduce the details of irregular pile body shape with high fidelity, and greatly improving the accuracy of contour measurement.

[0017] 3. This invention can adapt to different dust concentration conditions from normal to extreme, and has a wide range of applicability to various working conditions. The entire measurement process ensures high accuracy while taking efficiency into account, realizing fast, stable and accurate contour measurement and volume calculation of bulk cargo stacks in ports. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the overall structure of the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] As attached Figure 1 This paper presents a method for accurately measuring the outline of a bulk cargo yard at a port.

[0021] The following three sets of examples are derived: Example 1: Conventional dust coal pile working conditions.

[0022] This embodiment 1 focuses on a coal stockpile at a port, with a volume of 6000 m³ and a height of 22 m. The dust concentration in the working environment is 55 mg / m³. The contour measurement method for precise measurement of the stockpile's outline is employed in the port bulk cargo yard. The specific steps are as follows: S1. Establishment of the Primary Site Reference Coordinate System: A laser reflective target sphere with an accuracy of 0.015mm is embedded in the top of the fixed reference pile. Six non-collinear reference piles are arranged at adjacent intervals of 9m to form a rectangular primary site reference coordinate system. Positioning is performed using GPS / BeiDou dual-mode calibration, and the final scanning and recognition error is controlled within 2mm to ensure the stability and accuracy of the reference coordinate system.

[0023] S2. Secondary dynamic reference network layout: 65mm diameter magnetic wear-resistant reference spheres are used as temporary reference points with a surface roughness Ra controlled at 0.6μm. During layout, one temporary reference point is placed every 8m² in non-critical areas of the coal pile and on gentle slopes, and one temporary reference point is placed every 0.6m² in critical contour areas, including the top apex, bottom edge, and around slope inflection points, for a total of 82 temporary reference points. The straight-line distance between all temporary reference points and fixed reference piles is controlled at 17m, and the magnetic attraction property is used to ensure that the reference points do not shift during operation.

[0024] S3. Deployment and Calibration of Laser Scanning Equipment: Four 3D laser scanners were selected. The equipment has a ranging range of 0.5-100m, a single-point measurement accuracy of 0.16mm, and a scanning frequency of 400kHz. Following the principle of multi-station deployment, the four devices were arranged in open areas around the pile body. Each device covered three fixed reference piles and eight temporary reference points. The equipment was calibrated based on the primary site reference coordinate system. After calibration, the data synchronization error between the devices was controlled within 0.4mm.

[0025] S4. Layered scanning data acquisition: Layered scanning mode is adopted, with a scanning interval of 0.3m between each layer. A total of 74 layers are divided from the bottom to the top of the stack. The scanning time of a single station is controlled at 4min. The overlap rate of the scanning area of ​​adjacent stations is set at 40% to ensure no scanning blind spots. Since the current dust concentration is 55mg / m³, the dust removal and blowing device with a pressure of 0.6MPa is started to simultaneously complete the data acquisition of the stack outline, fixed benchmark piles and temporary reference points.

[0026] S5. Three-level reference verification and data correction: The least squares method is used to fit the characteristic curve of the pile surface. When selecting feature points, feature points are extracted within a radius of 0.6m in the neighborhood of the temporary reference point. 18 feature points are selected for each slope segment, including slope inflection points, top vertices, and bottom edge points, for a total of 144 feature points extracted. A quadratic polynomial surface model is used. The fitting process, including the meanings of each parameter and symbol, is as follows (fitting the scenario of this embodiment): Z is the vertical height of the three-dimensional coordinate elevation of a single feature point on the surface of the pile, which is the target value for the fitting calculation; X and Y are the three-dimensional plane coordinates (horizontal and vertical coordinates) of the same feature point, which are the input values ​​for the fitting calculation; Six fitting coefficients were obtained using the least squares method. Once the coefficients were determined, the surface model could be fixed to fit the actual surface morphology of the stack. An error equation was established and the coefficients were solved. After fitting, the residuals of each feature point were within 0.12 mm, indicating a satisfactory fit. The three-level reference verification standard was implemented: fixed reference pile coordinate deviation was 0.2 mm, temporary reference point coordinate deviation was 0.3 mm, feature point fitting deviation was 0.28 mm, and the outlier removal threshold was set at 0.6 mm. The final outlier percentage was 1.7%. Based on the fitted curve, the coordinates of the corresponding area scanning points were corrected, resulting in a data accuracy of 0.16 mm and a fluctuation value of 0.18 mm for multiple measurements.

[0027] S6. Point Cloud Data Filtering, Fusion, and 3D Reconstruction: Gaussian filtering algorithm is used to process point cloud data, and the Gaussian kernel function is constructed using a two-dimensional Gaussian function. , The meanings of each parameter and symbol are as follows (suitable for the scenario in this embodiment): The function output value is used to calculate the weight of a single point in the neighborhood of the point cloud, which determines the degree of influence of that point on the filtered coordinates of the center point. , It is the deviation of the planar coordinates of any point in the neighborhood from the center point of the neighborhood, which measures the spatial distance between the point and the center point. (Sigma) is the standard deviation (0.065m in this embodiment), which is the core parameter for adjusting the filtering effect and determines the width of the function curve; This is a normalization coefficient to ensure that the sum of the weights of all points in the neighborhood is 1, thus avoiding calculation bias. It is the natural constant (approximately 2.71828), forming an exponentially decaying term; This represents the exponential portion (which is negative overall), achieving a decay effect where "the farther the point is from the center point, the smaller its weight." Here, represents the standard. Poor balance between noise removal and detail preservation. , To calculate the planar coordinate deviation of points within the neighborhood relative to the center point, a weighted average is used. The weight of each point within the neighborhood is calculated based on a Gaussian function. After normalization of the sum of the weights, a weighted average is applied to the 3D coordinates (X, Y, Z) of each point within the neighborhood to obtain the filtered coordinates of the center point. The standard deviation σ in the Gaussian kernel function is set to 0.065m, the filtering radius is set to 0.06m, the data retention rate is 98.4%, and the point cloud stitching error is 0.32mm. The 3D contour is reconstructed through triangular mesh modeling with a mesh resolution of 0.058m and a contour detail restoration rate of 99.4%.

[0028] S7. Actual Measurement and Result Output: Twelve feature points in different areas were selected for actual measurement and verification. A handheld rangefinder with an accuracy of ±2mm was used. The average deviation was 0.2mm, which was deemed acceptable. The final calculation error of the coal pile volume was 0.22%. The entire measurement process took 27 minutes. The measurement error was controlled within the specified range, meeting the port coal pile measurement requirements.

[0029] Example 2: High dust ore stockpile conditions.

[0030] This Example 2 focuses on a port iron ore stockpile with a volume of 6800 m³ and a height of 30 m. The dust concentration in the working environment is 90 mg / m³. The specific measurement steps are as follows: S1. Layout of the primary site reference coordinate system: A laser reflective target ball with an accuracy of 0.012mm is embedded in the top of the fixed reference pile. Eight non-collinear reference piles are laid out with an adjacent spacing of 8.5m to form a polygonal primary reference coordinate system. After GPS / BeiDou dual-mode calibration, the scanning and recognition error is controlled within 1.5mm, providing a high-precision reference for subsequent measurements.

[0031] S2. Secondary dynamic reference network layout: Temporary reference points use magnetically attached wear-resistant reference balls with a diameter of 75mm and a surface roughness of Ra=0.5μm. The layout density is 1 point per 7m² in non-critical areas and 1 point per 0.7m² in critical contour areas, for a total of 105 temporary reference points. The straight-line distance between all temporary reference points and fixed reference piles is controlled within 18m, which is suitable for working conditions with high ore pile hardness and large loading and unloading impacts, and avoids reference point displacement.

[0032] S3. Deployment and Calibration of Laser Scanning Equipment: Eight 3D laser scanners were selected, with a ranging range of 0.5-100m, a single-point measurement accuracy of 0.17mm, and a scanning frequency of 380kHz. Multiple stations were deployed on high ground around the reactor body. Each device covered four fixed benchmark piles and nine temporary reference points. After calibration based on the first-level benchmark, the data synchronization error between devices was 0.35mm, ensuring the consistency of data collaboration among multiple devices.

[0033] S4. Layered scanning data acquisition: The layered scanning interval is set to 0.2m, and 150 layers are divided from bottom to top. The scanning time for a single station is 4.5min, and the overlap rate of the scanning areas of adjacent stations is 45%, which improves the data integrity of complex contour areas. Due to the dust concentration reaching 90mg / m³, the dust removal and purging device with a pressure of 0.8MPa is started to continuously purge the scanning path and simultaneously collect the contour of the pile body and the two-level reference point data to avoid data loss caused by dust obstruction.

[0034] S5. Three-level reference verification and data correction: The feature point selection radius is 0.7m, and 22 feature points are selected for each slope segment, for a total of 264 feature points. A quadratic polynomial surface model is used for fitting. The feature point residual after fitting is 0.14mm, which meets the qualified standard. Three-level reference verification: the coordinate deviation of the fixed benchmark pile is 0.15mm, the coordinate deviation of the temporary reference point is 0.35mm, the feature point fitting deviation is 0.32mm, the outlier removal threshold is 0.7mm, the outlier rate is 1.9%, the corrected data accuracy is 0.18mm, and the fluctuation value of multiple measurement data is 0.19mm, effectively offsetting the interference of dust on the reference points.

[0035] S6. Point cloud data filtering, fusion and 3D reconstruction: Gaussian filtering standard deviation σ=0.07m, filtering radius 0.07m, data retention rate 98.2%, point cloud stitching error 0.35mm, triangular mesh modeling mesh resolution 0.062m, 3D contour detail restoration degree 99.6%, clearly presenting the irregular curved surface shape of the ore pile.

[0036] S7. Actual Measurement and Result Output: 14 feature points were selected for actual measurement and verification. The handheld rangefinder had an accuracy of ±1.5mm, an average deviation of 0.25mm, a volume calculation error of 0.28%, and a total measurement time of 29 minutes. It is suitable for complex working conditions with large iron ore piles and high dust levels, and the measurement accuracy meets the port measurement requirements.

[0037] Example 3: Extreme dust-mixed bulk cargo stacking conditions.

[0038] This Example 3 focuses on a mixed bulk cargo stockpile of coal and ore at a port, with a volume of 5500 m³ and a height of 18 m. The dust concentration in the working environment is 110 mg / m³. The specific measurement steps are as follows: S1. Layout of the primary site reference coordinate system: A laser reflective target ball with an accuracy of 0.018mm is embedded in the top of the fixed reference pile. Five non-collinear reference piles are laid out with an adjacent spacing of 9.5m. After GPS / BeiDou dual-mode calibration, the scanning and identification error is 2.5mm to ensure the stability of the reference.

[0039] S2. Secondary dynamic reference network layout: Temporary reference points are magnetically attached wear-resistant reference spheres with a diameter of 58mm and a surface roughness Ra=0.7μm. There is one temporary reference point for every 9m² in non-critical areas and one temporary reference point for every 0.55m² in critical contour areas, for a total of 78 temporary reference points. The straight-line distance between the temporary reference points and the fixed reference piles is 16m. The magnetic design can resist collision interference during the loading and unloading of mixed bulk cargo.

[0040] S3. Deployment and calibration of laser scanning equipment: Six 3D laser scanners were selected, with a single-point measurement accuracy of 0.15mm and a scanning frequency of 420kHz. Each device covered three fixed benchmark piles and seven temporary reference points. The data synchronization error between devices was 0.45mm. Data reliability was improved through redundant acquisition of multiple devices.

[0041] S4. Layered scanning data acquisition: The layered scanning interval is 0.4m, divided into 45 layers, with a single station scanning time of 3.5min. The overlap rate of the scanning areas of adjacent stations is 50%, maximizing the coverage of irregular areas of the mixed pile. When the dust concentration is 110mg / m³, the dust removal and purging device with a pressure of 0.9MPa is started to remove dust from the scanning lens and reference point surface by high-frequency purging, ensuring the quality of data acquisition.

[0042] S5. Three-level reference verification and data correction: Feature point selection radius is 0.55m, 25 feature points are selected for each slope segment, and a total of 200 feature points are extracted. The residual after fitting the quadratic polynomial surface model is 0.13mm. Three-level reference verification: Fixed benchmark pile coordinate deviation is 0.25mm, temporary reference point coordinate deviation is 0.38mm, feature point fitting deviation is 0.3mm, outlier removal threshold is 0.75mm, outlier percentage is 1.8%, corrected data accuracy is 0.17mm, and the fluctuation value of multiple measurement data is 0.18mm. This achieves three-level reference linkage verification to offset the impact of extreme dust on reference points.

[0043] S6. Point cloud data filtering, fusion and 3D reconstruction: Gaussian filtering standard deviation σ=0.06m, filtering radius 0.055m, data retention rate 98.6%, point cloud stitching error 0.3mm, triangular mesh modeling mesh resolution 0.055m, 3D contour detail restoration degree 99.5%, accurately reconstructing the complex curved surface of the mixed bulk cargo stack.

[0044] S7. Actual Measurement and Result Output: 13 feature points were selected for actual measurement and verification. The handheld rangefinder had an accuracy of ±2.5mm, an average deviation of 0.22mm, a volume calculation error of 0.2%, and a total measurement time of 26 minutes. It still achieved high-precision measurement under extreme dust conditions, which is suitable for the measurement needs of mixed bulk cargo stacks in ports.

[0045] The following table is derived from the above three sets of embodiments: Table 1. Compliance Rate of Core Indicators in the Three-Level Reference Linkage Verification: Percentage calculation notes: 1. The core calculation formulas in the table are based on the following: Sub-item compliance rate (%) = (Number of compliance tests in this category / Total number of tests in this category) × 100%.

[0046] Fixed reference pile coordinate deviation compliance rate: The total number of tests is the total number of coordinate deviation tests of the fixed reference pile under the corresponding working conditions; the compliance test number is the number of tests in which the deviation value meets the preset accuracy threshold.

[0047] Temporary reference point coordinate deviation compliance rate: The total number of tests is the total number of times the coordinate deviation of the temporary reference point is detected under the corresponding working condition; the number of compliance tests is the number of times the deviation value meets the preset accuracy threshold.

[0048] Feature point fitting deviation compliance rate: The total number of detections is the total number of times the coordinate fitting deviation of the feature points of the stack is detected under the corresponding working condition; the number of compliance detections is the number of times the fitting deviation value meets the preset accuracy threshold.

[0049] The comprehensive compliance rate of the three-level linkage verification is calculated as follows: (compliance rate of fixed benchmark piles + compliance rate of temporary reference points + compliance rate of feature point fitting) / 3 × 100%, rounded to two decimal places.

[0050] Table 2: Reference Point Stability and Data Validity Ratio: Percentage calculation notes: 1. Percentage of temporary reference points with no displacement = Number of temporary reference points with no displacement / Total number of temporary reference points × 100%. In this embodiment, the magnetic design ensures no displacement, so the percentage is 100%.

[0051] 2. Percentage of valid feature points = (1 - Percentage of outliers) × 100%, directly calculated based on the outlier percentage data in the example.

[0052] 3. The point cloud data retention rate and contour detail restoration are directly taken from the actual measurement data of the embodiment, and the percentages are the direct presentation of the actual measurement results.

[0053] In summary, by establishing a three-level linkage verification mechanism of fixed site benchmarks, temporary dynamic reference points, and feature points on the surface of the stack, a complete spatial coordinate verification system was constructed. This system realizes real-time mutual verification and correction between multiple levels of references, effectively overcoming the inaccuracy of reference points caused by the mobility of bulk cargo, dust interference, and dynamic changes in operations. By realizing the linkage verification of three levels of references through site benchmarks, temporary dynamic references, and stack features, the reliability and anti-interference ability of the overall measurement are significantly improved, and the accuracy of contour measurement is greatly enhanced.

[0054] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for accurately measuring the outline of a bulk cargo storage yard at a port, characterized in that: The specific steps are as follows: S1. Establishment of the primary site reference coordinate system: Multiple non-collinear fixed reference stakes are set up to form a primary site reference coordinate system; S2. Secondary dynamic reference network deployment: Magnetic temporary reference points are deployed on the surface of the stack to form a secondary dynamic reference network; S3. Deployment and calibration of laser scanning equipment: Deploy laser scanning equipment at multiple stations and calibrate it based on a primary reference. S4. Layered scanning data acquisition: Layered scanning acquisition of the heap outline and two levels of reference point data; S5. Three-level reference verification and data correction: Through three-level reference verification, data is corrected by site benchmark, temporary reference and feature point; S6. Point cloud data filtering, fusion and 3D reconstruction: Filtering and fusing point cloud data and reconstructing the 3D contour; S7. Actual Measurement Verification and Result Output: Output the results after actual measurement verification, and keep the measurement error within a reduced range throughout the process.

2. The method for accurately measuring the outline of a bulk cargo yard loading and unloading structure according to claim 1, characterized in that: In S1, a laser reflective target ball is embedded at the top of the fixed reference pile, and the positioning is calibrated using GPS / BeiDou dual-mode.

3. The method for accurately measuring the outline of a bulk cargo yard loading and unloading structure according to claim 1, characterized in that: The temporary reference point in S2 is a magnetically attached wear-resistant reference ball, which is arranged at different densities in the non-critical areas and critical contour areas of the pile body, and maintains a reasonable straight-line distance from the fixed reference pile.

4. The method for accurately measuring the outline of a bulk cargo yard loading and unloading structure according to claim 1, characterized in that: The laser scanning equipment in S3 is a three-dimensional laser scanner. Multiple devices are deployed in a coordinated manner, with each device covering a certain number of fixed benchmark piles and temporary reference points to ensure data synchronization between devices.

5. The method for accurately measuring the outline of a bulk cargo yard loading and unloading structure according to claim 1, characterized in that: The S4 uses a layered scanning mode to control the overlap rate of the scanning areas of adjacent stations. When the dust concentration reaches a certain level, the dust removal and blowing device is activated.

6. The method for accurately measuring the outline of a bulk cargo yard loading and unloading structure according to claim 1, characterized in that: The S5 section sets a three-level reference verification standard, clarifies the threshold for outlier removal and the control requirements for the proportion of outliers, and ensures the accuracy of data correction.

7. The method for accurately measuring the outline of a bulk cargo yard loading and unloading structure according to claim 1, characterized in that: The S6 process uses a Gaussian filtering algorithm to process point cloud data and reconstructs the 3D contour through triangular mesh modeling, taking into account both noise removal and detail preservation, thus ensuring data validity and contour restoration accuracy.

8. The method for accurately measuring the outline of a bulk cargo yard loading and unloading structure according to claim 1, characterized in that: In step S7, feature points in different regions are selected for actual measurement and verification, and a handheld rangefinder is used to assist in the verification to ensure that the volume calculation error meets the requirements.

9. A method for accurately measuring the outline of a bulk cargo yard loading and unloading structure according to claim 1, characterized in that: The stockpile is for bulk cargo such as coal and ore, and is suitable for different dust concentration conditions at ports, enabling rapid and accurate measurement of a single stockpile.

10. A method for accurately measuring the outline of a bulk cargo yard loading and unloading structure according to claim 1, characterized in that: In S5, the least squares method is used to fit the characteristic curve of the pile surface. Through feature point screening, model construction, error equation establishment, coefficient solution and fitting verification and correction, the fitted curve is ensured to fit the measured feature points.

Images