Spatial distance measurement method and system based on three-dimensional measurement fusion technology

Abstract

The invention discloses a spatial distance measurement method and system based on a three-dimensional measurement fusion technology, and relates to the technical field of spatial distance measurement, and the method comprises the steps: arranging a sensor group which comprises a multi-view vision imaging module and a laser contourgraph module; performing multi-view image acquisition based on the multi-view vision imaging module to generate a preliminary depth map, and performing laser scanning based on the laser contourgraph module to generate three-dimensional point cloud data; the preliminary depth map and the three-dimensional point cloud data are subjected to data fusion, a three-dimensional model of a target object is generated, and the target object is a main bracket of the photo-thermal power generation trough type heat collector; and analyzing the three-dimensional model of the target object, determining the deviation value of each point position of the target object, and obtaining a space distance measurement result of the target object. The technical problem that in the prior art, space distance measurement of a main bracket of a photo-thermal power generation trough type heat collector is low in measurement precision is solved, and the technical effect of improving the measurement precision is achieved.

Description

Technical Field

[0001] The present invention relates to the field of space distance measurement technology, and in particular to a space distance measurement method and system based on three-dimensional measurement fusion technology. Background Art

[0002] With the rapid advancement of industrial production and technological development, solar thermal power generation technology has been widely used in the energy field. Among them, the main bracket of the trough collector is an important part of the solar thermal power generation system. The accuracy of its installation and maintenance directly affects the efficiency and stability of the entire system. However, with the expansion of production scale and the complexity of process flow, the spatial ranging process of the main bracket of the trough collector for solar thermal power generation faces increasing challenges. Traditional spatial ranging methods often have problems such as low measurement accuracy, which makes it difficult to meet the needs of modern solar thermal power generation for efficient, accurate and reliable measurement. Summary of the invention

[0003] The present application provides a spatial distance measurement method and system based on three-dimensional measurement fusion technology, which is used to solve the technical problem of low measurement accuracy in the spatial distance measurement of the main bracket of a solar thermal power generation trough collector in the prior art.

[0004] In view of the above problems, the present application provides a spatial ranging method and system based on three-dimensional measurement fusion technology.

[0005] In a first aspect of the present application, a spatial ranging method based on three-dimensional measurement fusion technology is provided, the method comprising:

[0006] A sensor group is deployed, wherein the sensor group includes a multi-view visual imaging module and a laser profiler module; multi-viewpoint image acquisition is performed based on the multi-view visual imaging module to generate a preliminary depth map, and laser scanning is performed based on the laser profiler module to generate three-dimensional point cloud data; the preliminary depth map is fused with the three-dimensional point cloud data to generate a three-dimensional model of a target object, wherein the target object is a main bracket of a solar thermal power generation trough collector; the three-dimensional model of the target object is analyzed to determine the deviation of each point of the target object, and obtain a spatial ranging result of the target object.

[0007] The second aspect of the present application provides a spatial ranging system based on three-dimensional measurement fusion technology, the system comprising:

[0008] A sensor group deployment unit, wherein the sensor group deployment unit deploys a sensor group, wherein the sensor group includes a multi-view visual imaging module and a laser profiler module; a data acquisition unit, wherein the data acquisition unit performs multi-view image acquisition based on the multi-view visual imaging module to generate a preliminary depth map, and performs laser scanning based on the laser profiler module to generate three-dimensional point cloud data; a data fusion unit, wherein the data fusion unit performs data fusion on the preliminary depth map and the three-dimensional point cloud data to generate a three-dimensional model of a target object, wherein the target object is a main bracket of a solar thermal power generation trough collector; and a deviation analysis unit, wherein the deviation analysis unit analyzes the three-dimensional model of the target object, determines the deviation of each point of the target object, and obtains a spatial ranging result of the target object.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0010] The present application deploys a sensor group, which includes a multi-eye vision imaging module and a laser profiler module; multi-viewpoint image acquisition is performed based on the multi-eye vision imaging module to generate a preliminary depth map, and laser scanning is performed based on the laser profiler module to generate three-dimensional point cloud data; the preliminary depth map and the three-dimensional point cloud data are fused to generate a three-dimensional model of the target object, which is the main bracket of the solar thermal power generation trough collector; the three-dimensional model of the target object is analyzed to determine the deviation of each point of the target object, and obtain the spatial distance measurement result of the target object. The present invention solves the technical problem of low measurement accuracy in the spatial distance measurement of the main bracket of the solar thermal power generation trough collector in the prior art, and achieves the technical effect of improving the measurement accuracy by integrating multi-eye vision imaging technology and laser profiler technology to monitor and accurately measure the target object in space in real time. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0012] Figure 1 A schematic diagram of a spatial distance measurement method flow based on three-dimensional measurement fusion technology provided in an embodiment of the present application;

[0013] Figure 2 A schematic diagram of the structure of a spatial ranging system based on three-dimensional measurement fusion technology provided in an embodiment of the present application.

[0014] Explanation of reference numerals: sensor group arrangement unit 11 , data acquisition unit 12 , data fusion unit 13 , deviation analysis unit 14 . DETAILED DESCRIPTION

[0015] The present application provides a spatial ranging method and system based on three-dimensional measurement fusion technology, aiming to solve the technical problem of low measurement accuracy in the spatial ranging of the main bracket of a solar thermal power generation trough collector in the prior art. By integrating multi-eye visual imaging technology and laser profilometer technology, the target objects in the space can be monitored and accurately measured in real time, thereby achieving the technical effect of improving measurement accuracy.

[0016] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0017] It should be noted that any variations of the terms "include" and "have" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or inherent to these processes, methods, products or devices.

[0018] Embodiment 1

[0019] like Figure 1 As shown, the present application provides a spatial ranging method based on three-dimensional measurement fusion technology, the method comprising:

[0020] Step S100: deploying a sensor group, wherein the sensor group includes a multi-eye vision imaging module and a laser profiler module.

[0021] Furthermore, step S100 in the method provided in the application embodiment also includes:

[0022] Based on the preset points, the multi-eye vision imaging module and the laser profiler module are deployed, the initial shooting angle of the sensor group is obtained, and shooting is performed to obtain test shooting data; based on the test shooting data, shooting blind spot identification is performed to determine whether there is a shooting blind spot; if so, the shooting angle of the sensor group is adjusted, and data is collected based on the adjusted sensor group to complete the deployment of the sensor group.

[0023] In the embodiment of the present application, the preset points are predetermined by technical experts based on the key support points and overall dimensions of the main bracket of the solar thermal power generation trough collector. These points will serve as the initial installation positions of the multi-eye visual imaging module and the laser profiler module to ensure that all key areas can be covered. After installing the multi-eye visual imaging module and the laser profiler module at the preset points, the initial shooting angle is obtained by adjusting according to the preset angle, and the first shooting is performed to obtain preliminary test shooting data.

[0024] Next, the shooting blind spots are identified based on the test shooting data. By analyzing the preliminarily shot multi-view images and three-dimensional point cloud data, any uncovered areas, i.e. shooting blind spots, are identified. Determine whether there are shooting blind spots. If so, adjust the shooting angle of the sensor group. Adjust the installation angle or position of the sensor group to eliminate the blind spots and ensure that all key areas can be effectively covered. Then, data is collected based on the adjusted sensor group, and shooting is performed again to obtain new test data. Repeat this process to check whether the data has shooting blind spots. Through multiple adjustments and data collection, ensure that all blind spots are eliminated and finally complete the deployment of the sensor group.

[0025] Step S200: performing multi-viewpoint image acquisition based on the multi-view imaging module to generate a preliminary depth map, and performing laser scanning based on the laser profiler module to generate three-dimensional point cloud data.

[0026] In an embodiment of the present application, first, a multi-view visual imaging module is started. The module is composed of multiple cameras, and the cameras are arranged at different viewpoints of the main bracket of the solar thermal power generation trough collector to cover the entire target area. Each camera captures images from different angles, and obtains multi-view image data through synchronous shooting. Next, the multi-view image is processed using a computer vision algorithm. The feature point extraction algorithm is applied to identify the key feature points in the image, and then the feature points of different viewpoints are matched using the image matching algorithm to generate the corresponding relationship of the feature points. Through triangulation, their positions in three-dimensional space are calculated based on the matched feature points to generate a preliminary depth map.

[0027] At the same time, the laser profiler module is started to perform laser scanning on the main bracket of the solar thermal power generation trough collector. The laser profiler accurately measures the three-dimensional coordinates of the surface of an object by emitting a laser beam and detecting its reflection time. The laser scanning covers all parts of the target object and generates high-precision three-dimensional point cloud data. The three-dimensional point cloud data consists of a large number of three-dimensional coordinate points, which record the shape and position of the surface of the target object in detail.

[0028] Furthermore, step S200 in the method provided in the application embodiment also includes:

[0029] Based on the multi-view imaging module, multi-viewpoint images of the main bracket of the solar thermal power generation trough collector are collected, and the camera is calibrated in real time using preset calibration points to obtain corrected image data; feature points are extracted from the corrected image data, and feature matching is performed to generate a set of matching feature points; calculations are performed based on the set of matching feature points to generate the preliminary depth map.

[0030] In the embodiment of the present application, during the acquisition process, the camera is calibrated in real time using the preset calibration points, and the position of the calibration points in the image is detected by the direct linear transformation method, and the displacement and angle deviation of the camera are corrected in real time to ensure the consistency and accuracy of the image data taken each time. The corrected image data eliminates the error caused by the change of the camera position, and obtains more accurate corrected image data.

[0031] Next, feature points are extracted from the corrected image data, and the scale-invariant feature transformation algorithm is used to extract key feature points in the corrected image data, such as corner points, edges and other significant features. Then, the normalized cross-correlation algorithm is used to match the feature points of different viewpoints to generate a set of matching feature points. The set of matching feature points contains the corresponding feature point pairs in different viewpoint images.

[0032] Finally, the depth value of each feature point in three-dimensional space is calculated by triangulation using the coordinate differences of the matched feature points. All calculated depth values ​​are combined to generate a preliminary depth map of the target object.

[0033] Step S300: fusing the preliminary depth map with the three-dimensional point cloud data to generate a three-dimensional model of a target object, wherein the target object is a main bracket of a solar thermal power generation trough collector.

[0034] In the embodiment of the present application, the preliminary depth map and the three-dimensional point cloud data are unified into the same coordinate system through a coordinate conversion algorithm. Then, the preliminary depth map and the three-dimensional point cloud data are matched with the feature point matching algorithm by combining the same feature points in the two sets of data. By extracting key feature points, such as corners and edges, and using a weighted average algorithm to fuse these matched feature point data, a three-dimensional model of the target object is generated.

[0035] Furthermore, step S300 in the method provided in the application embodiment also includes:

[0036] The preliminary depth map and the three-dimensional point cloud data are preprocessed to obtain a preliminary purified depth map and three-dimensional point cloud purified data; based on a coordinate transformation algorithm, the coordinate systems of the preliminary purified depth map and the three-dimensional point cloud purified data are unified; based on a key point matching algorithm, the key points of the preliminary purified depth map and the three-dimensional point cloud purified data are matched to obtain a key point matching set; the key point matching set is registered to obtain a precisely aligned depth map and a precisely aligned three-dimensional point cloud data; the precisely aligned depth map and the precisely aligned three-dimensional point cloud data are fused to obtain a three-dimensional model of the target object.

[0037] In the embodiment of the present application, a noise removal algorithm, such as Gaussian filtering, is first applied to the preliminary depth map to remove noise and redundant data in the image to obtain a preliminary purified depth map. At the same time, a statistical filter is applied to the three-dimensional point cloud data to obtain three-dimensional point cloud purified data by removing abnormal points and outliers.

[0038] Next, based on the coordinate transformation algorithm, the coordinate systems of the preliminary purified depth map and the 3D point cloud purified data are unified. For example, by applying methods such as homogeneous coordinate transformation, data from different sources are converted into a unified coordinate system. Then, the key feature points are extracted from the depth map and point cloud data using scale-invariant feature transformation or accelerated robust feature algorithm. These algorithms generate a set of key point descriptors with scale and rotation invariance by detecting local features in the image or point cloud. Next, a feature matching algorithm based on nearest neighbor search is applied to match these feature points to generate a set of key point matches.

[0039] Next, the iterative closest point algorithm is used to iteratively adjust the position and posture of the two sets of data by minimizing the Euclidean distance between matching feature points so that they can be accurately aligned in the same coordinate system, thereby obtaining a precisely aligned depth map and precisely aligned three-dimensional point cloud data.

[0040] Finally, the weighted average algorithm or surface fitting algorithm is used to fuse the aligned data, integrate the information of the depth map and point cloud data, and generate a three-dimensional model of the target object.

[0041] Furthermore, in the method provided in the embodiment of the application, the precisely aligned depth map and the precisely aligned three-dimensional point cloud data are fused to obtain the three-dimensional model of the target object, and further includes:

[0042] The precisely aligned depth map is interpolated to obtain a completed depth map; the precisely aligned three-dimensional point cloud data is density-homogenized to obtain homogenized three-dimensional point cloud data; the completed depth map and the homogenized three-dimensional point cloud data are fused by weighted averaging to obtain a preliminary fusion model; the preliminary fusion model is optimized by surface fitting to obtain a three-dimensional model of the target object.

[0043] In the embodiment of the present application, the missing or sparse data points in the precise alignment depth map are first interpolated by an interpolation algorithm such as bilinear interpolation or bicubic interpolation. These algorithms use the values ​​of surrounding known data points to estimate the depth value of the missing point, thereby generating a complete, hole-free completed depth map.

[0044] Next, the precisely aligned 3D point cloud data is subjected to density homogenization. The voxel grid filtering method is applied to divide the point cloud data into voxel grids of fixed size, and a representative point is used in each voxel grid, usually the center point of the voxel or the average value of all points in the voxel to represent the entire voxel. In this way, the density unevenness in the point cloud data is eliminated and homogenized 3D point cloud data is obtained.

[0045] Then, a weighted average algorithm is used to perform weighted fusion of the depth map and point cloud data according to the confidence and accuracy of each data source. Specifically, for each matching point, the weights from the depth map and point cloud data are calculated respectively, and the weighted average is performed to generate a preliminary fusion model, in which the weights of the depth map and point cloud data are pre-set.

[0046] Finally, a surface fitting algorithm, such as the thin plate spline method, is used to smooth and optimize the preliminary fusion model. The surface fitting algorithm eliminates noise and discontinuities in the data and generates a continuous and smooth three-dimensional surface model. Through these steps, the three-dimensional model of the target object is finally obtained.

[0047] Step S400: Analyze the three-dimensional model of the target object, determine the deviation of each point of the target object, and obtain the spatial distance measurement result of the target object.

[0048] Furthermore, step S400 in the method provided in the application embodiment also includes:

[0049] The three-dimensional model of the target object is gridded to obtain a gridded three-dimensional model; a comparative analysis is performed between a reference model and the gridded three-dimensional model to obtain preliminary deviation data; the preliminary deviation data is corrected based on an error compensation algorithm to obtain corrected deviation data; a statistical analysis is performed on the corrected deviation data to obtain a precise deviation amount of the target object; and the precise deviation amount of the target object is output as a spatial distance measurement result of the target object.

[0050] In an embodiment of the present application, a grid partitioning algorithm, such as Delaunay triangulation or tetrahedronization, is first applied to divide the three-dimensional model of the target object into a number of small units to form a regular gridded three-dimensional model. Next, a comparative analysis is performed based on the reference model and the gridded three-dimensional model to obtain preliminary deviation data. By applying an iterative closest point algorithm, the gridded three-dimensional model is aligned with the reference model so that the two are aligned in the same coordinate system. The Euclidean distance between each grid node and its corresponding reference node is then calculated to generate preliminary deviation data. The Kalman filter algorithm is then applied to analyze and correct the preliminary deviation data, eliminate random noise and systematic errors in the measurement, and obtain corrected deviation data.

[0051] Next, the statistical analysis method is used to calculate the average deviation and standard deviation of each grid node to obtain the precise deviation of the target object. Finally, the precise deviation of the target object is output as the spatial distance measurement result of the target object. At the same time, data visualization technology such as heat map or color mapping is used to intuitively display the deviation distribution on the 3D model and generate a detailed spatial distance measurement result report.

[0052] Furthermore, in the method provided in the embodiment of the application, the preliminary deviation data is corrected based on the error compensation algorithm to obtain the corrected deviation data, and the method also includes:

[0053] Environmental variables are introduced to construct a multidimensional compensation model, and the preliminary deviation data is compensated and corrected to obtain preliminary corrected deviation data; the preliminary corrected deviation data is iteratively optimized to obtain the corrected deviation data.

[0054] In the present application embodiment, first introduce environmental variables to build a multidimensional compensation model, the model is pre-built, when performing the multidimensional compensation model, first identify and record the environmental variables that affect the measurement accuracy, including temperature, humidity and light intensity, etc., these variables have a significant impact on the measurement results. By repeatedly measuring under different environmental conditions, collect preliminary deviation data and corresponding environmental variable values, ensure that the data covers various possible ranges of environmental variables. Take environmental variables as input features, preliminary deviation data as output labels, and perform data preprocessing to ensure that all input features are on the same scale. Next, select a suitable machine learning algorithm, such as support vector machine to build a compensation model. The collected data set is divided into a training set and a validation set, and the model is trained on the training set so that it learns the relationship between environmental variables and preliminary deviation data, and selects the optimal model parameters by cross-validation and hyperparameter optimization. Test the model on the validation set, evaluate its prediction performance, and ensure that the model has good generalization ability. Obtain a multidimensional compensation model through the above process.

[0055] Next, the preliminary deviation data are corrected using a multidimensional compensation model to eliminate the systematic errors caused by environmental factors, thereby obtaining preliminary corrected deviation data.

[0056] Finally, an iterative optimization algorithm is used, such as the iterative closest point ICP algorithm, which calculates the error between the current correction data and the ideal value through initial iterations, and adjusts the correction parameters in each iteration to minimize the error between the measured data and the benchmark model. The ICP algorithm calculates the distance between the current deviation data and the target value through continuous iterations, and adjusts the correction parameters so that the two gradually approach each other. The error minimization step is repeated until the error of the correction data reaches the preset convergence condition, and finally the correction deviation data is obtained.

[0057] In the embodiments of the present application, in summary, the embodiments of the present application have at least the following technical effects:

[0058] The present application deploys a sensor group, which includes a multi-eye vision imaging module and a laser profiler module; multi-viewpoint image acquisition is performed based on the multi-eye vision imaging module to generate a preliminary depth map, and laser scanning is performed based on the laser profiler module to generate three-dimensional point cloud data; the preliminary depth map and the three-dimensional point cloud data are fused to generate a three-dimensional model of the target object, which is the main bracket of the solar thermal power generation trough collector; the three-dimensional model of the target object is analyzed to determine the deviation of each point of the target object, and obtain the spatial distance measurement result of the target object. The present invention solves the technical problem of low measurement accuracy in the spatial distance measurement of the main bracket of the solar thermal power generation trough collector in the prior art, and achieves the technical effect of improving the measurement accuracy by integrating multi-eye vision imaging technology and laser profiler technology to monitor and accurately measure the target object in space in real time.

[0059] Embodiment 2

[0060] Based on the same inventive concept as the spatial distance measurement method based on three-dimensional measurement fusion technology in the aforementioned embodiment, Figure 2 As shown, the present application provides a spatial distance measurement system based on three-dimensional measurement fusion technology, and the system and method embodiments in the embodiments of the present application are based on the same inventive concept. The system includes:

[0061] A sensor group deployment unit 11, wherein the sensor group deployment unit 11 deploys a sensor group, wherein the sensor group includes a multi-eye vision imaging module and a laser profiler module;

[0062] A data acquisition unit 12, wherein the data acquisition unit 12 performs multi-view image acquisition based on the multi-view imaging module to generate a preliminary depth map, and performs laser scanning based on the laser profiler module to generate three-dimensional point cloud data;

[0063] A data fusion unit 13, wherein the data fusion unit 13 performs data fusion on the preliminary depth map and the three-dimensional point cloud data to generate a three-dimensional model of a target object, wherein the target object is a main bracket of a solar thermal power generation trough collector;

[0064] The deviation analysis unit 14 analyzes the three-dimensional model of the target object, determines the deviation of each point of the target object, and obtains the spatial distance measurement result of the target object.

[0065] Furthermore, the system is also used to implement the following functions:

[0066] Based on the preset points, the multi-eye vision imaging module and the laser profiler module are arranged, the initial shooting angle of the sensor group is obtained, and shooting is performed to obtain test shooting data;

[0067] Based on the test shooting data, performing shooting blind area identification to determine whether there is a shooting blind area;

[0068] If so, the shooting angle of the sensor group is adjusted, and data is collected based on the adjusted sensor group to complete the deployment of the sensor group.

[0069] Furthermore, the system is also used to implement the following functions:

[0070] Based on the multi-view imaging module, multi-viewpoint images of the main bracket of the solar thermal power generation trough collector are collected, and the camera is calibrated in real time using preset calibration points to obtain calibrated image data;

[0071] Extracting feature points from the corrected image data, and performing feature matching to generate a set of matching feature points;

[0072] Calculation is performed based on the set of matching feature points to generate the preliminary depth map.

[0073] Furthermore, the system is also used to implement the following functions:

[0074] Preprocessing the preliminary depth map and the three-dimensional point cloud data to obtain a preliminary purified depth map and three-dimensional point cloud purified data;

[0075] Unifying the coordinate systems of the preliminary cleansed depth map and the three-dimensional point cloud cleansed data based on a coordinate transformation algorithm;

[0076] Based on a key point matching algorithm, matching the key points of the preliminary cleansed depth map and the three-dimensional point cloud cleansed data to obtain a key point matching set;

[0077] Registering the key point matching set to obtain a precisely aligned depth map and precisely aligned three-dimensional point cloud data;

[0078] The precisely aligned depth map and the precisely aligned three-dimensional point cloud data are fused to obtain the three-dimensional model of the target object.

[0079] Furthermore, the system is also used to implement the following functions:

[0080] Performing interpolation processing on the precisely aligned depth map to obtain a completed depth map;

[0081] Performing density homogenization processing on the precisely aligned three-dimensional point cloud data to obtain homogenized three-dimensional point cloud data;

[0082] Fusing the completed depth map and the homogenized three-dimensional point cloud data by weighted averaging to obtain a preliminary fusion model;

[0083] The preliminary fusion model is optimized by surface fitting to obtain the three-dimensional model of the target object.

[0084] Furthermore, the system is also used to implement the following functions:

[0085] Meshing the three-dimensional model of the target object to obtain a meshed three-dimensional model;

[0086] Comparative analysis is performed based on the benchmark model and the gridded three-dimensional model to obtain preliminary deviation data;

[0087] Correcting the preliminary deviation data based on an error compensation algorithm to obtain corrected deviation data;

[0088] Performing statistical analysis on the correction deviation data to obtain an accurate deviation amount of the target object;

[0089] The precise deviation of the target object is output as the spatial distance measurement result of the target object.

[0090] Furthermore, the system is also used to implement the following functions:

[0091] Introducing environmental variables to construct a multidimensional compensation model, performing compensation correction on the preliminary deviation data, and obtaining preliminary corrected deviation data;

[0092] The preliminary correction deviation data is iteratively optimized to obtain the correction deviation data.

[0093] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order and continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0094] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0095] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A spatial distance measurement method based on three-dimensional measurement fusion technology, characterized in that: The method comprises: Arrange a sensor group, wherein the sensor group includes a multi-eye vision imaging module and a laser profiler module; Perform multi-view image acquisition based on the multi-view imaging module to generate a preliminary depth map, and perform laser scanning based on the laser profiler module to generate three-dimensional point cloud data; The preliminary depth map is fused with the three-dimensional point cloud data to generate a three-dimensional model of a target object, wherein the target object is a main bracket of a solar thermal power generation trough collector; The three-dimensional model of the target object is analyzed to determine the deviation of each point of the target object and obtain the spatial distance measurement result of the target object.

2. The spatial distance measurement method based on three-dimensional measurement fusion technology according to claim 1, characterized in that: A sensor group is arranged, wherein the sensor group includes a multi-eye vision imaging module and a laser profiler module, including: Based on the preset points, the multi-eye vision imaging module and the laser profiler module are arranged, the initial shooting angle of the sensor group is obtained, and shooting is performed to obtain test shooting data; Based on the test shooting data, performing shooting blind area identification to determine whether there is a shooting blind area; If so, the shooting angle of the sensor group is adjusted, and data is collected based on the adjusted sensor group to complete the deployment of the sensor group.

3. The spatial distance measurement method based on three-dimensional measurement fusion technology according to claim 1, characterized in that: Based on the multi-view imaging module, multi-view image acquisition is performed to generate a preliminary depth map, including: Based on the multi-view imaging module, multi-viewpoint images of the main bracket of the solar thermal power generation trough collector are collected, and the camera is calibrated in real time using preset calibration points to obtain calibrated image data; Extracting feature points from the corrected image data, and performing feature matching to generate a set of matching feature points; Calculation is performed based on the set of matching feature points to generate the preliminary depth map.

4. The spatial distance measurement method based on three-dimensional measurement fusion technology according to claim 1, characterized in that: The preliminary depth map is fused with the three-dimensional point cloud data to generate a three-dimensional model of the target object, including: Preprocessing the preliminary depth map and the three-dimensional point cloud data to obtain a preliminary purified depth map and three-dimensional point cloud purified data; Unifying the coordinate systems of the preliminary cleansed depth map and the three-dimensional point cloud cleansed data based on a coordinate transformation algorithm; Based on a key point matching algorithm, matching the key points of the preliminary cleansed depth map and the three-dimensional point cloud cleansed data to obtain a key point matching set; Registering the key point matching set to obtain a precisely aligned depth map and precisely aligned three-dimensional point cloud data; The precisely aligned depth map and the precisely aligned three-dimensional point cloud data are fused to obtain the three-dimensional model of the target object.

5. The spatial distance measurement method based on three-dimensional measurement fusion technology according to claim 4, characterized in that: The precisely aligned depth map and the precisely aligned three-dimensional point cloud data are fused to obtain the three-dimensional model of the target object, including: Performing interpolation processing on the precisely aligned depth map to obtain a completed depth map; Performing density homogenization processing on the precisely aligned three-dimensional point cloud data to obtain homogenized three-dimensional point cloud data; Fusing the completed depth map and the homogenized three-dimensional point cloud data by weighted averaging to obtain a preliminary fusion model; The preliminary fusion model is optimized by surface fitting to obtain the three-dimensional model of the target object.

6. The spatial distance measurement method based on three-dimensional measurement fusion technology according to claim 1, characterized in that: Analyzing the three-dimensional model of the target object, determining the deviation of each point of the target object, and obtaining the spatial distance measurement result of the target object, including: Meshing the three-dimensional model of the target object to obtain a meshed three-dimensional model; Comparative analysis is performed based on the benchmark model and the gridded three-dimensional model to obtain preliminary deviation data; Correcting the preliminary deviation data based on an error compensation algorithm to obtain corrected deviation data; Performing statistical analysis on the correction deviation data to obtain an accurate deviation amount of the target object; The precise deviation of the target object is output as the spatial distance measurement result of the target object.

7. The spatial distance measurement method based on three-dimensional measurement fusion technology according to claim 6, characterized in that: Correcting the preliminary deviation data based on an error compensation algorithm to obtain corrected deviation data includes: Introducing environmental variables to construct a multidimensional compensation model, performing compensation correction on the preliminary deviation data, and obtaining preliminary corrected deviation data; The preliminary correction deviation data is iteratively optimized to obtain the correction deviation data.

8. The spatial distance measurement system based on three-dimensional measurement fusion technology is characterized by: The system comprises: A sensor group deployment unit, wherein the sensor group deployment unit deploys a sensor group, wherein the sensor group includes a multi-eye vision imaging module and a laser profiler module; A data acquisition unit, wherein the data acquisition unit performs multi-view image acquisition based on the multi-view imaging module to generate a preliminary depth map, and performs laser scanning based on the laser profiler module to generate three-dimensional point cloud data; A data fusion unit, wherein the data fusion unit fuses the preliminary depth map with the three-dimensional point cloud data to generate a three-dimensional model of a target object, wherein the target object is a main bracket of a solar thermal power generation trough collector; The deviation analysis unit analyzes the three-dimensional model of the target object, determines the deviation of each point of the target object, and obtains the spatial distance measurement result of the target object.

Images