Multifunctional automatic space point location tracking measurement method and device

Abstract

The invention discloses a multifunctional automatic space point location tracking measurement method and device, and relates to the technical field of tracking measurement, and the method comprises the steps: obtaining N monitoring points, carrying out the priority ranking, obtaining the monitoring point sequence information, presetting a monitoring collection frequency, and generating the monitoring region information according to the monitoring point sequence information; a built-in high-precision navigation module automatically navigates to monitoring area information, a target icon is fixed on a target monitoring point, and target monitoring image information of the monitoring area information is obtained; training and constructing a point location tracking module, performing target positioning and coordinate extraction, and determining target pixel coordinate information; and target coordinate calculation is carried out to obtain actual space coordinates of the target, the actual space coordinates of the target are stored in sequence, and N space point location measurement results are generated. The technical problems that the monitoring frequency is difficult to guarantee, the measurement efficiency is low and the positioning precision is insufficient in the prior art are solved, and the technical effect of improving the efficiency and precision of space point location tracking measurement is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of tracking measurement, and particularly relates to a multi-functional automatic spatial point position tracking measurement method and device. Background Art

[0002] During the process of engineering construction, the monitoring of the spatial coordinates of key points is of great significance for ensuring project quality and structural safety. With the continuous improvement of the level of industrial automation, the demand for high-frequency, high-precision, and continuous monitoring of spatial points is increasing day by day, and the limitations of traditional measurement methods are becoming increasingly prominent. Existing spatial point position measurement methods mostly rely on manual operation or semi-automatic equipment. Common methods include manual reading or using a total station for point position observation. Such methods not only rely highly on the technical level of operators, have complex operation processes, but also require multi-person cooperation, resulting in relatively low overall measurement efficiency. When facing the continuous measurement tasks of multiple monitoring points, traditional methods are difficult to balance the timeliness and accuracy requirements of data collection, especially insufficient in high-frequency monitoring scenarios, and it is difficult to achieve continuous and automatic monitoring of points. Summary of the Invention

[0003] This application provides a multi-functional automatic spatial point position tracking measurement method and device for solving the technical problems of difficult-to-guarantee monitoring frequency, low measurement efficiency, and insufficient positioning accuracy in the prior art.

[0004] In view of the above problems, this application provides a multi-functional automatic spatial point position tracking measurement method and device.

[0005] In the first aspect of this application, a multi-functional automatic spatial point position tracking measurement method is provided. The method includes: Obtain N monitoring points, perform priority sorting on the N monitoring points to obtain monitoring point sequence information, input the theoretical coordinate information and preset error range of the target monitoring point in sequence according to the monitoring point sequence information according to the preset monitoring and acquisition frequency, and generate monitoring area information according to the theoretical coordinate information and preset error range; the built-in high-precision navigation module automatically navigates to the monitoring area information, fixes a target icon at the target monitoring point, and uses the target icon as a reference to collect the target monitoring image information of the monitoring area information through a high-precision telescope; train and construct a point position tracking module, use the point position tracking module to perform target positioning and coordinate extraction on the target monitoring image information, and determine the target pixel coordinate information; perform target coordinate calculation based on the telescope camera parameters and the target pixel coordinate information to obtain the actual spatial coordinates of the target, and store the actual spatial coordinates of the target in sequence to generate N spatial point position measurement results.

[0006] The second aspect of the present application provides a multi-functional automated spatial point position tracking and measuring device, which includes: A monitoring area information acquisition module, configured to acquire N monitoring points, perform a priority ranking on the N monitoring points to obtain monitoring point sequence information, input the theoretical coordinate information and a preset error range of the target monitoring points in sequence according to the monitoring point sequence information at a preset monitoring acquisition frequency, and generate monitoring area information according to the theoretical coordinate information and the preset error range; A monitoring image information acquisition module, configured to automatically navigate to the monitoring area information by means of a built-in high-precision navigation module, fix a target icon at the target monitoring point, and use the target icon as a reference to collect and obtain the target monitoring image information of the monitoring area information through a high-precision telescope; A coordinate information confirmation module, configured to train and construct a point position tracking module, use the point position tracking module to perform target positioning and coordinate extraction on the target monitoring image information, and determine the target pixel coordinate information; A point position measurement result confirmation module, configured to perform target coordinate calculation based on the telescope camera parameters and the target pixel coordinate information to obtain the actual spatial coordinates of the target, and store the actual spatial coordinates of the target in sequence to generate N spatial point position measurement results.

[0007] One or more technical solutions provided in the present application have at least the following technical effects or advantages: The present application acquires N monitoring points, performs a priority ranking on the N monitoring points to obtain monitoring point sequence information, inputs the theoretical coordinate information and a preset error range of the target monitoring points in sequence according to the monitoring point sequence information at a preset monitoring acquisition frequency, and generates monitoring area information according to the theoretical coordinate information and the preset error range; automatically navigates to the monitoring area information by means of a built-in high-precision navigation module, fixes a target icon at the target monitoring point, and uses the target icon as a reference to collect and obtain the target monitoring image information of the monitoring area information through a high-precision telescope; trains and constructs a point position tracking module, uses the point position tracking module to perform target positioning and coordinate extraction on the target monitoring image information, and determines the target pixel coordinate information; performs target coordinate calculation based on the telescope camera parameters and the target pixel coordinate information to obtain the actual spatial coordinates of the target, and stores the actual spatial coordinates of the target in sequence to generate N spatial point position measurement results. The present invention solves the technical problems of difficult-to-guarantee monitoring frequency, low measurement efficiency, and insufficient positioning accuracy in the prior art. By adopting a combination of automatic navigation, image recognition, and spatial coordinate calculation, it performs high-frequency and automated recognition and positioning on multiple monitoring points, achieving the technical effects of improving the efficiency and accuracy of spatial point position tracking and measurement. Description of the Drawings

[0008] To more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the accompanying drawings required for the description of the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0009] Figure 1 Schematic flowchart of a multifunctional automated spatial point tracking measurement method provided by an embodiment of the present application; Figure 2 Schematic structural diagram of a multifunctional automated spatial point tracking measurement device provided by an embodiment of the present application.

[0010] Explanation of reference numerals: Monitoring area information acquisition module 11, monitoring image information acquisition module 12, coordinate information confirmation module 13, point measurement result confirmation module 14. Detailed implementation manners

[0011] The present application provides a multifunctional automated spatial point tracking measurement method and device. To solve the technical problems of difficult-to-guarantee monitoring frequency, low measurement efficiency, and insufficient positioning accuracy in the prior art, by combining automatic navigation, image recognition, and spatial coordinate calculation, high-frequency and automated identification and positioning of multiple monitoring points are performed, achieving the technical effects of improving the efficiency and accuracy of spatial point tracking measurement.

[0012] The following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of them. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts belong to the scope of protection of the present application.

[0013] It should be noted that any variations of the terms "including" and "having" are intended to cover non-exclusive inclusion. For example, a process, method, device, product, or server including a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or modules not clearly listed or inherent to these processes, methods, products, or devices.

[0014] Embodiment 1, as Figure 1 shown, the present application provides a multifunctional automated spatial point tracking measurement method, and the method includes: Step S100: Obtain N monitoring points, perform priority sorting on the N monitoring points to obtain monitoring point sequence information, and input the theoretical coordinate information and preset error range of the target monitoring points in sequence according to the monitoring point sequence information at a preset monitoring acquisition frequency. Generate monitoring area information based on the theoretical coordinate information and the preset error range.

[0015] In the embodiments of the present application, first, N preset monitoring points are obtained, and a priority factor evaluation system is constructed based on preset monitoring point priority factors such as criticality, urgency, and monitoring difficulty. Through the comprehensive evaluation of each monitoring point, N sets of priority factor coefficients are obtained, and then the monitoring points are sorted to generate ordered monitoring point sequence information. The above N monitoring points include key points in the engineering project that are sensitive to structural changes or have a greater impact on safety, such as bridge bearings, top of high piers, top of deep foundation pit slopes, deformation joint nodes, construction cable anchor points, etc. These points carry key structural stress or displacement information and are important objects for engineering quality control and risk warning.

[0016] After obtaining the monitoring point sequence information, a preset monitoring acquisition frequency is further set. This acquisition frequency is dynamically adjusted by technical experts according to the priority of the monitoring points. For example, higher sampling frequencies are assigned to points with higher criticality or more frequent changes to increase their monitoring density. Subsequently, in the order of the monitoring point sequence information, the theoretical coordinate information and preset error range of each target monitoring point are input in sequence. The theoretical coordinate information is used to determine the ideal position of the target point in space, while the preset error range is used to define the allowable spatial offset limit.

[0017] Based on the theoretical coordinates and the error range, a three-dimensional cube-shaped monitoring area boundary is constructed with the theoretical coordinates as the center and the error range as the radius, and a suitable grid type and grid density are set according to the measurement accuracy requirements. The boundary is spatially divided and reconstructed to form a monitoring area model. Finally, in combination with the actual distribution characteristics of the monitoring points and the surrounding environmental conditions, the model is verified and optimized, and the final monitoring area information is output. The above entire process is executed by a tracking measurement robot, and the tracking measurement robot automatically completes the loading and parsing of monitoring point information and the scheduling setting of the sampling frequency.

[0018] Further, in the method provided by the application embodiments, the obtaining of the monitoring point sequence information further includes: Preset monitoring point priority factors, where the monitoring point priority factors include criticality, urgency, and monitoring difficulty; according to the monitoring point priority factors, establish a priority factor evaluation system, where the priority factor evaluation system includes specific evaluation indicators for each monitoring point priority factor; according to the priority factor evaluation system, conduct a priority evaluation on the N monitoring points to obtain N priority factor coefficient sets; based on the N priority factor coefficient sets, conduct a priority ranking on the N monitoring points to obtain monitoring point sequence information.

[0019] In the embodiment of the present application, first preset multiple core factors for judging the importance of monitoring points, including criticality, urgency, and monitoring difficulty. Among them, criticality is used to reflect the importance of the monitoring point in the entire structure or measurement, urgency indicates whether there is a rapid change or potential risk at the point that needs to be monitored first, and monitoring difficulty is used to evaluate the complexity of stably and accurately measuring this point in the actual environment.

[0020] To enable the monitoring point priority factors to have a quantifiable evaluation ability, establish a priority factor evaluation system. This system refines criticality, urgency, and monitoring difficulty into multiple specific evaluation indicators such as structure impact factor, historical change rate, alarm frequency, recognition success rate, and occlusion degree respectively, to ensure that each factor has a clear quantitative basis. Based on the Analytic Hierarchy Process (AHP), assign weights to each priority factor to construct a priority factor weight matrix. When using the AHP, first construct a pairwise comparison judgment matrix of the priority factors, compare the relative importance of every two factors in the overall evaluation. For example, set the importance of criticality relative to urgency to 3, indicating that criticality is more important, and the importance of urgency relative to monitoring difficulty to 2, and construct a complete comparison matrix accordingly; then normalize this matrix and calculate the average value of each row to obtain the weight coefficients of each factor and form a weight vector. Conduct a consistency test on this matrix to ensure the rationality of the weight assignment and the consistency of the judgment matrix, so that the priority evaluation results have stability and credibility.

[0021] After the construction of the priority factor evaluation system is completed, the priority evaluation is performed on each of the N monitoring points one by one. Specifically, the data of each monitoring point in each evaluation index dimension is collected. For example, the structure influence factor of a certain monitoring point is 0.85, the daily average displacement rate is 2.1 mm / d, the occlusion degree is 30%, and the recognition success rate is 92%. The collected index data is normalized, and the original data is uniformly mapped to the range of 0 to 1 according to the interval. For example, the normalized score is calculated by the difference between the maximum value and the minimum value, so as to eliminate the dimensional difference and numerical scale difference between different indexes. After the index data of all monitoring points is normalized, it will be weighted and summed with the priority factor weight matrix to obtain the comprehensive priority score of each monitoring point. Through this process, a set of priority factor coefficients including all monitoring points is generated, where each monitoring point corresponds to a set of numerical combinations reflecting its criticality, urgency and monitoring difficulty and the final weighted total score.

[0022] After the construction of the priority factor coefficient set is completed, further sorting analysis is carried out based on the N priority factor coefficient sets. First, the normalized evaluation indexes of each monitoring point are organized by dimension to generate N priority coefficient matrices, which are used to uniformly represent the distribution of all points in the dimensions of criticality, urgency and monitoring difficulty. Subsequently, based on the existing priority factor evaluation system, index data mining is carried out to construct a priority factor evaluation index data set, which is used to enhance the understanding and modeling of the correlation between indexes. On this basis, the principal component analysis (PCA) method is applied to the index data set for dimension denoising and feature extraction to obtain the principal component load coefficients of each priority factor, complete the PCA weight allocation of the priority factors, and form an optimized priority factor weight matrix. The above-mentioned N priority coefficient matrices and the PCA weight matrix are subjected to matrix weighted operation to calculate the monitoring priority coefficient of each monitoring point, which represents its comprehensive importance under the multi-factor fusion. Finally, the N monitoring points are sorted in descending order according to the monitoring priority coefficient, and the monitoring point sequence information is output, which is used to guide the subsequent measurement scheduling and frequency configuration.

[0023] After the priority evaluation of all monitoring points is completed, all monitoring points are sorted according to the total scores obtained from the priority factor coefficient set, and a monitoring point sequence information is formed by arranging them from high to low. This monitoring point sequence information not only clarifies the execution order of subsequent measurement tasks, but also can be directly mapped to the monitoring acquisition frequency to realize the optimization of the sampling strategy based on priority. For example, the monitoring points with a priority score higher than 0.8 are set to high-frequency sampling (such as once every 5 minutes), those with a score between 0.6 and 0.8 are set to medium-frequency sampling (such as once every 15 minutes), and those lower than 0.6 are set to low-frequency sampling (once every 30 minutes), so as to ensure that resources are focused on key points and improve the efficiency and accuracy of the overall measurement response.

[0024] Further, in the method provided by the application embodiment, obtaining the monitoring point sequence information further includes: Generating N priority coefficient matrices according to the N sets of priority factor coefficients; performing index data mining based on the priority factor evaluation system to obtain a priority factor evaluation index data set; performing PCA weight assignment on the monitoring point priority factors according to the priority factor evaluation index data set to generate a priority factor weight matrix; performing weighted calculation on the N priority coefficient matrices and the priority factor weight matrix to obtain N monitoring priority coefficients, and performing priority ranking on the N monitoring points based on the N monitoring priority coefficients to obtain the monitoring point sequence information.

[0025] In the embodiment of the present application, first, according to the obtained N sets of priority factor coefficients, N priority coefficient matrices are generated by using a matrix reconstruction method. Specifically, the normalized scores of each monitoring point in the three dimensions of criticality, urgency, and monitoring difficulty are filled into a structured two-dimensional matrix in a unified order. The rows of this matrix represent the monitoring point numbers, and the columns correspond to the priority factors, forming an N×3 numerical matrix, which is convenient for unified linear operations in subsequent calculations.

[0026] After obtaining the N priority coefficient matrices, based on the priority factor evaluation system, index data mining is performed on the existing monitoring point evaluation data, and the Pearson correlation analysis method is used to identify the statistical correlation between different priority factors. This method determines the degree of linear correlation by calculating the Pearson correlation coefficient between any two columns of data (i.e., two factors), thereby clarifying which factors have a high common explanatory ability and which factors are relatively independent. For example, in the above matrix, if the correlation coefficient between criticality and urgency is 0.85, it indicates a high correlation between the two, while the correlation between monitoring difficulty and the other factors is weak, indicating that it may represent an independent dimensional feature.

[0027] Immediately afterwards, principal component analysis (PCA) is applied to the priority factor evaluation index data set for feature extraction and dimension optimization. In PCA analysis, first, a covariance matrix is constructed for the priority coefficient matrix, and the principal component vectors are solved by eigenvalue decomposition; the load value of the principal component represents the projection weight of each original evaluation factor in the direction of this principal component. Therefore, the principal component that can explain the largest variance is extracted according to the load value size and normalized, and finally a priority factor weight matrix is formed. For example, among the above three factors, if the first principal component can explain 80% of the variance and its corresponding normalized load value is (0.6, 0.5, 0.3), then this vector will be used as the final weight assignment for comprehensive score calculation.

[0028] Subsequently, matrix multiplication is used to perform linear weighted calculation on the aforementioned priority factor weight matrix and N priority coefficient matrices. That is, the inner product operation is performed between a row of priority score vectors corresponding to each monitoring point and the weight vector to obtain a scalar value, which is used as the final monitoring priority coefficient of that point. For example, the score vector of point A is (0.9, 0.7, 0.4), which is multiplied by the weight vector (0.6, 0.5, 0.3) and summed to obtain its priority coefficient as 0.9×0.6 + 0.7×0.5 + 0.4×0.3 = 1.01. The same applies to other monitoring points, and the calculation of N priority coefficients is completed.

[0029] Finally, based on the obtained N monitoring priority coefficients, a sorting operation is performed. The quicksort algorithm is used to sort all monitoring points in descending order of the priority coefficient to form monitoring point sequence information.

[0030] Furthermore, in the method provided by the application embodiment, the generating of the monitoring area information further includes: Taking the theoretical coordinate information as the center and the preset error range as the radius, the monitoring area boundary is determined. The monitoring area boundary is a three-dimensional cubic area; according to the area measurement requirements and model complexity, the grid type and grid density are determined; the monitoring area boundary is spatially divided and reconstructed according to the grid type and grid density to generate a monitoring area model; the monitoring area model is verified and optimized based on the actual distribution of monitoring points and environmental conditions, and the generated monitoring area information is output.

[0031] In the embodiment of the present application, first, with the theoretical coordinate information of each monitoring point as the center and based on the preset error range, a three-dimensional monitoring boundary is constructed. This process uses the spatial bounding box generation method, taking the theoretical coordinate as the center point of the bounding box and the error range as the boundary extension value in the three-dimensional direction. It expands outward by ± error values in the x, y, and z directions respectively to generate the smallest bounding cube. This cube is defined as the monitoring area boundary to ensure that the subsequent recognition searches for target points within a reasonable range, reducing the misrecognition rate and computational redundancy. For example, when the theoretical coordinate is (1.0, 2.0, 1.5) meters and the error range is set to 0.3 meters, the constructed cube boundary range is x ∈ [0.7, 1.3], y ∈ [1.7, 2.3], z ∈ [1.2, 1.8].

[0032] After determining the boundary, in combination with the requirements for measurement accuracy and model complexity in the current monitoring area, the empirical threshold method is used to determine the grid type and grid density. The specific approach is as follows: Select the cubic grid as the spatial discretization structure according to the target accuracy level, and automatically set the granularity of the unit space division based on the grid density standard matching the accuracy requirement in the empirical library. For example, in the centimeter-level measurement accuracy scenario, the system will set the grid density to one cell per 0.01 meters; if it is millimeter-level accuracy, it will be set to one cell per 0.001 meters, so as to ensure that the grid modeling can meet the spatial resolution requirements of downstream coordinate calculation and image recognition.

[0033] After obtaining the boundary and grid parameters, a regular grid division algorithm is used to reconstruct the spatial division of the monitoring area boundary. This algorithm generates a set of three-dimensional voxel units by equally dividing the x, y, and z axes of the cubic region. Each voxel represents the basic component in the monitoring area model. Calculate the spatial coordinate range of each unit in turn, and establish a complete three-dimensional grid model based on this, so that the entire monitoring area is accurately divided into regular sub-regions with clear numbers, providing spatial support for subsequent image acquisition perspective adjustment and path planning.

[0034] After completing the construction of the preliminary grid model, the monitoring area model is verified and optimized based on the model evaluation algorithm. This process uses the on-site image comparison method, that is, calls the telescope to collect the images of the current area, maps and projects the monitoring area model into the image coordinate system, and compares it with the target point positions in the actual image to determine whether the spatial range of the model covers the real target points. When boundary deviation, recognition deviation, or spatial occlusion is found, adjust the grid boundary range or re-divide the density parameters until the model highly coincides with the on-site image data. Finally, the output monitoring area information is the area structure after three-dimensional modeling and image verification and optimization.

[0035] Step S200: The built-in high-precision navigation module automatically navigates to the monitoring area information, fixes the target icon at the target monitoring point, and takes the target icon as a reference to collect the target monitoring image information of the monitoring area information through a high-precision telescope.

[0036] In the embodiment of the present application, after the tracking and measurement robot obtains the monitoring area information of the target monitoring point, it first starts the automatic navigation process through its built-in high-precision navigation module and goes to the monitoring area where the target monitoring point is located. This navigation module combines GPS positioning and attitude sensors (such as gyroscopes, accelerometers) to achieve navigation control, plans the travel route according to the theoretical coordinates of the monitoring point, controls the robot to move precisely along the set path, and corrects the direction in real time to ensure that it smoothly enters the target monitoring area.

[0037] After reaching the monitoring area, the robot fixes the target icon at the target position. The icon is an easily recognizable marking pattern, such as a black-and-white contrast obvious grid mark or an identification symbol with coding (such as AprilTag), which can be accurately located through an image recognition algorithm. The icon is automatically attached or placed on the surface near the monitoring point by the robot's own robotic arm or mounting device to ensure its stable visibility in the image.

[0038] After the target icon is arranged, the robot starts the high-precision telescope it carries to take pictures of the monitoring area. The telescope includes a high-resolution camera and an adjustable lens. By controlling the lens direction and focal length, it is aligned with the target monitoring area. After focusing and collecting the image, the image should completely contain the target icon and the surrounding environmental features for subsequent image analysis and coordinate calculation.

[0039] The finally obtained image data is the target monitoring image information of the monitoring area information, which contains a clear target pattern, surrounding structures, time information and positioning information at the time of shooting.

[0040] Step S300: Train and construct a point tracking module, use the point tracking module to perform target positioning and coordinate extraction on the target monitoring image information, and determine the target pixel coordinate information.

[0041] In the embodiment of the present application, first train and construct a point tracking module. Based on a preset target icon, collect image samples containing the target to form an initial target feature data set, and perform image enhancement processing on this data set, including operations such as rotation, scaling, brightness adjustment, and occlusion simulation, to generate a more diverse extended target feature data set. Subsequently, use a convolutional neural network (CNN) to train the extended data set to generate a target recognition sub-network specifically used to recognize the target icon in the image. On this basis, further construct a target tracking sub-network for processing the dynamic position change of the target in continuous image frames. This sub-network shares weights with the recognition sub-network, that is, uses the same feature extraction layer to reduce model redundancy and improve tracking robustness. Finally, fuse and connect the recognition sub-network and the tracking sub-network to construct a complete point tracking module.

[0042] Input the obtained target monitoring image information into the point tracking module. The module automatically processes the image, locates the specific position of the target icon in the image, extracts and outputs the two-dimensional coordinate value of the target center, that is, the target pixel coordinate information.

[0043] Furthermore, in the method provided by the embodiment of the application, the training and construction of the point tracking module further includes: Collect feature data based on the target icon to obtain a target feature data set, and perform image data augmentation on the target feature data set to obtain an augmented target feature data set; use a convolutional neural network to perform recognition training and verification optimization on the augmented target feature data set to generate a target recognition sub-network; according to the target recognition sub-network, construct a target tracking sub-network, and the target tracking sub-network and the target recognition sub-network are shared weight networks; merge and connect the target tracking sub-network and the target recognition sub-network to construct the point position tracking module.

[0044] In the embodiment of the present application, first, based on the used target icon, the target images under different angles, illuminations, distances, and background conditions are collected by a camera, and an initial target feature data set is established by using image acquisition technology. During the collection process, typical environmental scenarios are selected, such as indoors, outdoors, strong light, and occlusion conditions, and the target is imaged multiple times to ensure that the collected images can fully cover various variation forms in actual applications and provide comprehensive sample support for subsequent model training. After obtaining the preliminary image samples, an image enhancement method is used to augment the target feature data set. The specific operations include using image processing tools such as OpenCV to perform standard enhancement operations on the images, such as rotation (e.g., within the range of ±30 degrees), scaling (e.g., ±20%), brightness adjustment, adding noise, etc., and introducing processing methods such as occlusion simulation and background interference superposition to form an augmented target feature data set with rich pose changes and environmental interference features, so that the training model has the ability to maintain recognition robustness in complex scenarios.

[0045] Subsequently, a convolutional neural network (CNN) is used to perform recognition training on the augmented data set to construct a target recognition sub-network. This process uses a typical object detection architecture, such as YOLO, to extract image features such as edges and corners in multiple convolutional layers by inputting the image and the labeled target position information, and combines the classification layer and the regression layer to output the category and position information of the target in the image. During the training process, the cross-entropy loss function is used to optimize the classification results, and the IoU loss function is used to adjust the positioning bounding box to ensure that the network can stably output the accurate position of the target in the image.

[0046] After the recognition sub-network training is completed, a correlation filtering tracking method is used to construct a target tracking sub-network for continuously tracking the target in a video stream or consecutive images. This method calculates the response of the image region based on the correlation filter, determines the maximum response position by sliding a template region in the current frame image and matching it with the target feature template in the previous frame image, so as to achieve the position tracking of the target in the time series. To improve consistency, the tracking sub-network and the recognition sub-network share the weight network, that is, both use the same convolutional feature extraction structure to achieve unity at the parameter level to ensure the consistency and efficiency of the recognition and tracking results.

[0047] Finally, the trained target recognition sub-network and the target tracking sub-network are structurally docked, and a complete point position tracking module is constructed through network structure splicing and output fusion methods.

[0048] Furthermore, in the method provided by the application embodiment, obtaining the augmented target feature data set further includes: Obtain image enhancement factor conditions, where the image enhancement factor conditions include lighting conditions, background complexity, and target icon pose changes; design change parameters for each factor condition in the image enhancement factor conditions to generate a target image change matrix set; based on the target image change matrix set, sequentially perform image enhancement and augmentation on the target feature data set to obtain the augmented target feature data set.

[0049] In the embodiment of the present application, to improve the recognition ability of the point position tracking module for target icons in complex environments, systematic image enhancement processing is performed on the initially collected target feature data set to construct a training sample set with diversity and robustness. Specifically, first, by analyzing the possible change factors in the image acquisition process in the actual monitoring scenario, the image enhancement factor conditions are obtained, mainly including three categories: lighting conditions, background complexity, and target icon pose changes. Among them, lighting conditions refer to the brightness and contrast changes caused by natural light or artificial light sources during image acquisition, background complexity reflects the interference degree of the area other than the target in the image, such as the presence of debris, line textures, etc., and target icon pose changes include angle and size transformations such as rotation, scaling, and skew that the icon may undergo in actual applications.

[0050] For each of the above enhancement factors, change parameter design is carried out, that is, a numerical change range and rules are established for each factor to form a controllable model for target image changes. For example, in terms of lighting conditions, the brightness adjustment range is set to ±30%, and the contrast adjustment range is set to ±20%; in terms of background complexity, combinations of adding different textures or noise layers are designed; in terms of pose changes, parameters such as the icon rotation angle of ±45°, the scaling ratio of 80% - 120%, and the affine distortion angle not exceeding ±10° are set. The above parameter change combinations form a multi-dimensional change reference structure through a regular arrangement method, and these parameters are encoded to form a target image change matrix set, where each row in the matrix represents a complete parameter configuration of a group of image enhancement operations.

[0051] After the matrix set construction is completed, image enhancement and augmentation are performed on the initial target feature data set based on this set of transformation matrices. In specific operations, the transformation combinations in the transformation matrix are applied to each image in the original data set in sequence. For example, for an image, first adjust the brightness, then perform rotation, and then add background texture interference, thus generating multiple different versions of enhanced images. This processing process is executed batch by an image processing library (such as OpenCV), and finally a diversified image set that is multiplied and expanded for the entire original data set is formed and output as an augmented target feature data set.

[0052] Step S400: Calculate the target coordinates based on the telescope camera parameters and the target pixel coordinate information, obtain the actual spatial coordinates of the target, and store the actual spatial coordinates of the target in sequence to generate N spatial point measurement results.

[0053] In the embodiment of the present application, accurate coordinate calculation is performed based on the telescope camera parameters and the extracted target pixel coordinate information. Specifically, first, the parameters of the camera used by the telescope are analyzed, and the key parameters for imaging and coordinate transformation are extracted, including the internal parameter matrix (such as focal length, principal point position, etc.), distortion coefficients (including radial distortion and tangential distortion), and the external parameter matrix (representing the rotation and translation relationship between the camera coordinate system and the world coordinate system). The above parameters can be obtained at one time through existing camera calibration methods such as Zhang Zhengyou calibration method and are configured during the system initialization phase.

[0054] After obtaining the camera parameters, image distortion correction is performed on the target pixel coordinates extracted from the image based on the distortion coefficients to eliminate the geometric distortion caused by the non-ideal characteristics of the lens during the imaging process. The corrected coordinates are the corrected target pixel coordinates, representing the position of the target in the ideal distortion-free imaging model. Subsequently, according to the internal parameter matrix, this corrected coordinate is transformed and mapped, and through the pinhole camera model or projection transformation method, it is converted into the three-dimensional coordinates in the camera coordinate system, that is, the target camera three-dimensional coordinates, reflecting the position relationship between the target point and the camera.

[0055] Then, the external parameter matrix is used to perform a spatial transformation on the camera three-dimensional coordinates to convert them from the camera coordinate system to the unified world coordinate system to obtain the final actual spatial coordinates of the target.

[0056] When the image recognition and coordinate calculation of all monitoring points are completed, each actual spatial coordinate of the target is stored in an orderly manner according to the monitoring point sequence information, and finally N spatial point measurement results are generated.

[0057] Furthermore, in the method provided by the application embodiment, obtaining the actual spatial coordinates of the target further includes: Extract the coordinate influence parameters of the telescope camera parameters to determine the internal parameter matrix, distortion coefficients, and external parameter matrix; correct the radial and tangential distortions of the target pixel coordinate information based on the distortion coefficients to obtain the corrected target pixel coordinates; perform conversion mapping on the corrected target pixel coordinates according to the internal parameter matrix to obtain the three-dimensional coordinates of the target camera; perform conversion mapping on the three-dimensional coordinates of the target camera based on the external parameter matrix to determine the actual spatial coordinates of the target.

[0058] In the embodiment of the present application, first, a calibration operation is performed on the telescope camera to extract the coordinate influence parameters for coordinate conversion, including the internal parameter matrix, distortion coefficients, and external parameter matrix. The internal parameter matrix is mainly obtained by taking a series of images with known structures using a camera calibration board (such as a checkerboard) and automatically calculating using vision tools such as OpenCV to obtain parameters such as the focal length and the position of the principal point, which are used to describe the relationship between pixels and the geometric center of the image. The distortion coefficients are also extracted simultaneously during the calibration process, usually including three radial distortion coefficients and two tangential distortion coefficients, which are used for subsequent correction of image distortion. The external parameter matrix is obtained by collecting images of reference points at different positions in space by the camera and calibrating the position in combination with a total station or a laser rangefinder to obtain the spatial orientation and position of the camera.

[0059] After obtaining the above parameters, start processing the identified target pixel coordinate information. First, based on the extracted distortion coefficients, use the image undistortion method to correct the radial and tangential distortions of the pixel points. For example, a target point that appears at the edge of the image, with its original pixel position being (1450, 980), after radial distortion correction, the position may be adjusted to (1422, 970), and then after tangential distortion correction, it finally becomes (1410, 965), forming the corrected target pixel coordinates. This correction process accurately restores the geometric deviation caused by the lens in the image by reconstructing the imaging geometric relationship, which is a key link in image coordinate accuracy control.

[0060] After the correction is completed, according to the extracted internal parameter matrix, by obtaining the horizontal focal length, vertical focal length, and principal point coordinates of the camera, and combining the target depth coordinates determined by the laser ranging module or known structure dimensions, perform normalization processing on the corrected pixel coordinates to obtain the normalized two-dimensional coordinates of the target in the camera's field of view. Subsequently, multiply the normalized coordinates by the target depth coordinates respectively to obtain the two-dimensional coordinates of the target camera, and then combine them with the depth coordinates to finally obtain the three-dimensional coordinates of the target camera of the target in the camera coordinate system.

[0061] Next, use the external parameter matrix of the camera to extract the rotation matrix and translation vector therein, and convert the aforementioned three-dimensional camera coordinates into homogeneous coordinate format. Multiply the rotation matrix by the homogeneous coordinates and then sum with the translation vector to complete the spatial transformation from the camera coordinate system to the world coordinate system, and finally calculate the actual spatial coordinates of the target in the global measurement environment.

[0062] Further, in the method provided by the application embodiment, obtaining the three-dimensional camera coordinates of the target further includes: According to the internal parameter matrix, obtain the horizontal focal length, vertical focal length, and horizontal principal point coordinates, and at the same time obtain the target depth coordinates; based on the horizontal focal length, vertical focal length, and horizontal principal point coordinates, normalize the corrected target pixel coordinates to obtain normalized target pixel coordinates; take the product of the target depth coordinates and the normalized target pixel coordinates as the two-dimensional camera coordinates of the target, and combine the two-dimensional camera coordinates of the target and the target depth coordinates to obtain the three-dimensional camera coordinates of the target.

[0063] In the embodiment of the present application, first extract the horizontal focal length, vertical focal length, and horizontal principal point coordinates from the internal parameter matrix of the camera. These parameters are obtained through a standard camera calibration process. For example, using a checkerboard image sequence and using Zhang Zhengyou's calibration method to reverse calculate the focal length and principal point to ensure that the parameters have sufficient geometric accuracy. Taking a set of conventional calibration results as an example, set the horizontal focal length to 1000 pixels, the vertical focal length to 1000 pixels, and the horizontal principal point coordinates to (960, 540). Then, obtain the target depth coordinates corresponding to the position of this target point. This depth information is accurately measured by a laser rangefinder (such as a TOF laser sensor) or directly provided by the structure data in a scene where the target height is known. Based on this example, set the target depth coordinates to 5.0 meters.

[0064] After preparing the internal and depth data, based on the horizontal focal length, vertical focal length, and horizontal principal point coordinates, normalize the corrected target pixel coordinates. Assume that the corrected target pixel coordinates are (u = 1250, v = 980), and the normalization calculation process is x = (1250 - 960) / 1000 = 0.29, y = (980 - 540) / 1000 = 0.44. Through normalization, the normalized target pixel coordinates are obtained as (0.29, 0.44), that is, the image coordinates are standardized to a relative coordinate system aligned with the camera optical axis direction for the next calculation.

[0065] Subsequently, take the product of the target depth coordinates and the normalized target pixel coordinates as the two-dimensional camera coordinates of the target, that is, X = 0.29×5.0 = 1.45 meters, Y = 0.44×5.0 = 2.20 meters, indicating the spatial positions of the target point in the horizontal and vertical directions. Accordingly, the two-dimensional camera coordinates of the target are obtained as (1.45, 2.20) meters.

[0066] Finally, the target camera 2D coordinates and target depth coordinates are combined to obtain the target camera 3D coordinates, which is (1.45, 2.20, 5.0) meters. The coordinate result is the target camera 3D coordinates, which indicates the exact position of the target in the 3D space coordinate system with the camera as the origin.

[0067] Furthermore, in the method provided in the embodiment of the application, the determining of the actual spatial coordinates of the target further includes: According to the external parameter matrix, the rotation matrix and the translation vector are obtained, and the three-dimensional coordinates of the target camera are converted into target homogeneous coordinates; based on the rotation matrix and the translation vector, the target homogeneous coordinates are multiplied and summed to determine the actual spatial coordinates of the target.

[0068] In an embodiment of the present application, first, the system obtains the rotation matrix and translation vector according to the extrinsic matrix. The extrinsic matrix is ​​used to describe the position and posture information of the camera relative to the world coordinate system. The system collects multiple target images of known positions during the camera calibration phase, and uses a single perspective point estimation method (PnP) to extract the spatial correspondence between the image points and the world coordinate points, thereby calculating the rotation matrix and the corresponding translation vector. The rotation matrix is ​​used to describe the directional transformation of the camera around each axis in space, and the translation vector represents the displacement of the camera coordinate origin in the world coordinate system.

[0069] Secondly, the target camera 3D coordinates are converted into target homogeneous coordinates. To achieve unified spatial transformation operations, the coordinate extension method is used to add a fixed value to the original 3D coordinates to form a 4D coordinate expression. This coordinate representation method is widely used in the field of 3D computer vision and can integrate rotation and translation operations into a linear transformation process, making subsequent processing more concise and unified in structure.

[0070] Then, the target homogeneous coordinates are multiplied and summed based on the rotation matrix and translation vector. In this step, the rotation matrix and translation vector are spliced ​​to form a complete spatial transformation structure, and it is calculated item by item with the target homogeneous coordinates. The specific method is to substitute the target coordinates into each set of rotation coefficients and add the corresponding translation value to obtain the horizontal, vertical and height positions of the target in the global three-dimensional space.

[0071] Finally, the three coordinate values ​​in the actual space are extracted from the calculation results to determine the actual space coordinates of the target.

[0072] In the embodiments of the present application, in summary, the embodiments of the present application have at least the following technical effects: This application obtains N monitoring points, sorts the priorities of the N monitoring points to obtain monitoring point sequence information, and inputs the theoretical coordinate information and preset error range of the target monitoring points in sequence according to the monitoring point sequence information at the preset monitoring acquisition frequency. According to the theoretical coordinate information and the preset error range, monitoring area information is generated; the built-in high-precision navigation module automatically navigates to the monitoring area information, fixes a target icon at the target monitoring point, and uses the target icon as a reference to collect the target monitoring image information of the monitoring area information through a high-precision telescope; a point position tracking module is trained and constructed, and the point position tracking module is used to perform target positioning and coordinate extraction on the target monitoring image information to determine the target pixel coordinate information; target coordinate calculation is performed based on the telescope camera parameters and the target pixel coordinate information to obtain the actual spatial coordinates of the target, and the actual spatial coordinates of the target are stored in sequence to generate N spatial point position measurement results. The present invention solves the technical problems of difficult-to-guarantee monitoring frequency, low measurement efficiency, and insufficient positioning accuracy in the prior art. By adopting a combination of automatic navigation, image recognition, and spatial coordinate calculation, high-frequency and automated recognition and positioning of multiple monitoring points are performed, achieving the technical effects of improving the efficiency and accuracy of spatial point position tracking measurement.

[0073] Embodiment 2, based on the same inventive concept as the multifunctional automated spatial point position tracking measurement method in the foregoing embodiment, as Figure 2 shown, this application provides a multifunctional automated spatial point position tracking measurement device. The device in the embodiment of this application and the method embodiment are based on the same inventive concept. Among them, the device includes: A monitoring area information acquisition module 11, configured to obtain N monitoring points, sort the priorities of the N monitoring points to obtain monitoring point sequence information, and input the theoretical coordinate information and preset error range of the target monitoring points in sequence according to the monitoring point sequence information at the preset monitoring acquisition frequency. According to the theoretical coordinate information and the preset error range, monitoring area information is generated; a monitoring image information acquisition module 12, configured to use the built-in high-precision navigation module to automatically navigate to the monitoring area information, fix a target icon at the target monitoring point, and use the target icon as a reference to collect the target monitoring image information of the monitoring area information through a high-precision telescope; a coordinate information confirmation module 13, configured to train and construct a point position tracking module, and use the point position tracking module to perform target positioning and coordinate extraction on the target monitoring image information to determine the target pixel coordinate information; a point position measurement result confirmation module 14, configured to perform target coordinate calculation based on the telescope camera parameters and the target pixel coordinate information to obtain the actual spatial coordinates of the target, and store the actual spatial coordinates of the target in sequence to generate N spatial point position measurement results.

[0074] Further, the device is also used to implement the following functions: Preset monitoring point priority factors, where the monitoring point priority factors include criticality, urgency, and monitoring difficulty; according to the monitoring point priority factors, establish a priority factor evaluation system, where the priority factor evaluation system includes specific evaluation indicators for each monitoring point priority factor; according to the priority factor evaluation system, conduct a priority evaluation on the N monitoring points to obtain N sets of priority factor coefficients; based on the N sets of priority factor coefficients, conduct a priority ranking on the N monitoring points to obtain monitoring point sequence information.

[0075] Further, the device is also used to implement the following functions: According to the N sets of priority factor coefficients, generate N priority coefficient matrices; based on the priority factor evaluation system, conduct index data mining to obtain a priority factor evaluation index data set; according to the priority factor evaluation index data set, conduct PCA weight allocation on the monitoring point priority factors to generate a priority factor weight matrix; perform weighted calculation on the N priority coefficient matrices and the priority factor weight matrix to obtain N monitoring priority coefficients, and based on the N monitoring priority coefficients, conduct a priority ranking on the N monitoring points to obtain the monitoring point sequence information.

[0076] Further, the device is also used to implement the following functions: Taking the theoretical coordinate information as the center and the preset error range as the radius, determine the monitoring area boundary, where the monitoring area boundary is a three-dimensional cubic area; according to the area measurement requirements and model complexity, determine the grid type and grid density; according to the grid type and grid density, conduct spatial division and reconstruction on the monitoring area boundary to generate a monitoring area model; based on the actual distribution of monitoring points and environmental conditions, verify and optimize the monitoring area model, and output and generate the monitoring area information.

[0077] Further, the device is also used to implement the following functions: Based on the target icon, collect feature data to obtain a target feature data set, and perform image data enhancement on the target feature data set to obtain an expanded target feature data set; use a convolutional neural network to conduct recognition training and verification optimization on the expanded target feature data set to generate a target recognition sub-network; according to the target recognition sub-network, construct a target tracking sub-network, where the target tracking sub-network and the target recognition sub-network are shared weight networks; merge and connect the target tracking sub-network and the target recognition sub-network to construct the point position tracking module.

[0078] Further, the device is also used to implement the following functions: Obtain image enhancement factor conditions, where the image enhancement factor conditions include illumination conditions, background complexity, and target icon pose changes; design change parameters for each factor condition in the image enhancement factor conditions respectively to generate a set of target image change matrices; sequentially perform image enhancement and expansion on the target feature data set based on the set of target image change matrices to obtain an expanded target feature data set.

[0079] Further, the device is also used to implement the following functions: Extract coordinate influence parameters from the telescope camera parameters to determine the internal parameter matrix, distortion coefficient, and external parameter matrix; perform radial distortion and tangential distortion correction on the target pixel coordinate information based on the distortion coefficient to obtain corrected target pixel coordinates; perform transformation mapping on the corrected target pixel coordinates according to the internal parameter matrix to obtain the three-dimensional coordinates of the target camera in the camera; perform transformation mapping on the three-dimensional coordinates of the target camera in the camera based on the external parameter matrix to determine the actual spatial coordinates of the target.

[0080] Further, the device is also used to implement the following functions: According to the internal parameter matrix, obtain the horizontal focal length, vertical focal length, and horizontal principal point coordinates, and at the same time obtain the depth coordinates of the target; perform normalization processing on the corrected target pixel coordinates based on the horizontal focal length, vertical focal length, and horizontal principal point coordinates to obtain normalized target pixel coordinates; use the product of the depth coordinates of the target and the normalized target pixel coordinates as the two-dimensional coordinates of the target camera in the camera, and combine the two-dimensional coordinates of the target camera in the camera and the depth coordinates of the target to obtain the three-dimensional coordinates of the target camera in the camera.

[0081] Further, the device is also used to implement the following functions: According to the external parameter matrix, obtain the rotation matrix and translation vector, and at the same time convert the three-dimensional coordinates of the target camera in the camera into homogeneous coordinates of the target; perform product summation calculation on the homogeneous coordinates of the target based on the rotation matrix and translation vector to determine the actual spatial coordinates of the target.

[0082] It should be noted that the above sequence of embodiments of the present application is only for description and does not represent the superiority or inferiority of the embodiments. And the above describes specific embodiments of this specification. The processes depicted in the 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.

[0083] The above are only the preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included within the protection scope of the present application.

[0084] This specification and the accompanying drawings are merely illustrative of the present application and are considered to cover any and all modifications, variations, combinations, or equivalents within the scope of the present application. Obviously, those skilled in the art can make various changes and modifications 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 equivalent technologies, the present application is intended to include these changes and modifications.

Claims

1. A multifunctional automated spatial point tracking measurement method, characterized in that: The method comprises: Acquire N monitoring points, prioritize the N monitoring points, obtain monitoring point sequence information, preset monitoring acquisition frequency, input theoretical coordinate information and preset error range of target monitoring points in sequence according to the monitoring point sequence information, and generate monitoring area information according to the theoretical coordinate information and the preset error range; The built-in high-precision navigation module automatically navigates to the monitoring area information, fixes a target icon on the target monitoring point, and uses the target icon as a reference to acquire target monitoring image information of the monitoring area information through a high-precision telescope; Training and constructing a point tracking module, using the point tracking module to locate and extract coordinates of the target monitoring image information, and determining the target pixel coordinate information; The target coordinates are calculated based on the telescope camera parameters and the target pixel coordinate information to obtain the actual space coordinates of the target, and the actual space coordinates of the target are stored in sequence to generate N spatial point measurement results.

2. A multifunctional automated spatial point tracking and measurement method as claimed in claim 1, characterized in that: The obtaining of monitoring point sequence information includes: Preset monitoring point priority factors, wherein the monitoring point priority factors include criticality, urgency, and monitoring difficulty; Establishing a priority factor evaluation system based on the priority factors of the monitoring points, wherein the priority factor evaluation system includes specific evaluation indicators of the priority factors of each monitoring point; Performing priority evaluation on the N monitoring points according to the priority factor evaluation system to obtain N priority factor coefficient sets; The N monitoring points are prioritized based on the N priority factor coefficient sets to obtain monitoring point sequence information.

3. A multifunctional automated spatial point tracking measurement method as claimed in claim 2, characterized in that: The prioritizing the N monitoring points based on the N priority factor coefficient sets to obtain monitoring point sequence information includes: Generating N priority coefficient matrices according to the N priority factor coefficient sets; Performing indicator data mining based on the priority factor evaluation system to obtain a priority factor evaluation indicator data set; Performing PCA weight allocation on the priority factors of the monitoring points according to the priority factor evaluation index data set to generate a priority factor weight matrix; The N priority coefficient matrices and the priority factor weight matrix are weighted to obtain N monitoring priority coefficients, and the N monitoring points are prioritized based on the N monitoring priority coefficients to obtain the monitoring point sequence information.

4. A multifunctional automated spatial point tracking and measurement method as claimed in claim 2, characterized in that: The generating of monitoring area information comprises: Taking the theoretical coordinate information as the center and the preset error range as the radius, the monitoring area boundary is determined, and the monitoring area boundary is a three-dimensional cube area; Determine the grid type and density based on regional measurement requirements and model complexity; Performing spatial division and reconstruction on the monitoring area boundary according to the grid type and grid density to generate a monitoring area model; The monitoring area model is verified and optimized based on the actual distribution of monitoring points and environmental conditions, and the monitoring area information is generated as an output.

5. A multifunctional automated spatial point tracking and measurement method as claimed in claim 1, characterized in that: The training and construction point tracking module includes: Performing feature data collection based on the target icon to obtain a target feature data set, and performing image data enhancement on the target feature data set to obtain an expanded target feature data set; Using a convolutional neural network to perform recognition training and verification optimization on the expanded target feature data set to generate a target recognition subnetwork; According to the target recognition subnetwork, a target tracking subnetwork is constructed, wherein the target tracking subnetwork and the target recognition subnetwork are shared weight networks; The target tracking subnetwork and the target recognition subnetwork are merged and connected to construct the point tracking module.

6. A multifunctional automated spatial point tracking and measurement method as claimed in claim 5, characterized in that: The expanded target feature data set is obtained, including: Acquiring image enhancement factor conditions, wherein the image enhancement factor conditions include lighting conditions, background complexity, and target icon posture changes; Designing the variation parameters of each element condition in the image enhancement element condition respectively to generate a target image variation matrix set; Based on the target image change matrix set, the target feature data set is sequentially enhanced and expanded to obtain an expanded target feature data set.

7. A multifunctional automated spatial point tracking and measurement method as claimed in claim 1, characterized in that: The step of obtaining the actual spatial coordinates of the target includes: Extracting coordinate influencing parameters of the telescope camera parameters to determine an internal parameter matrix, a distortion coefficient, and an external parameter matrix; Performing radial distortion and tangential distortion correction on the target pixel coordinate information based on the distortion coefficient to obtain corrected target pixel coordinates; Transforming and mapping the correction target pixel coordinates according to the internal reference matrix to obtain the target camera three-dimensional coordinates; The three-dimensional coordinates of the target camera are transformed and mapped based on the external parameter matrix to determine the actual spatial coordinates of the target.

8. A multifunctional automated spatial point tracking and measurement method as claimed in claim 7, characterized in that: The step of obtaining the three-dimensional coordinates of the target camera includes: According to the internal parameter matrix, the horizontal focal length, the vertical focal length and the horizontal principal point coordinates are obtained, and the target depth coordinates are obtained at the same time; Based on the horizontal focal length, the vertical focal length and the horizontal principal point coordinates, normalizing the correction target pixel coordinates to obtain normalized target pixel coordinates; The product of the target depth coordinate and the normalized target pixel coordinate is used as the target camera two-dimensional coordinate, and the target camera two-dimensional coordinate and the target depth coordinate are combined to obtain the target camera three-dimensional coordinate.

9. A multifunctional automated spatial point tracking and measurement method as claimed in claim 7, characterized in that: Determining the actual spatial coordinates of the target includes: According to the external parameter matrix, a rotation matrix and a translation vector are obtained, and at the same time, the three-dimensional coordinates of the target camera are converted into target homogeneous coordinates; The target homogeneous coordinates are multiplied and summed based on the rotation matrix and the translation vector to determine the actual space coordinates of the target.

10. A multifunctional automated spatial point tracking and measuring device, characterized in that: The device comprises: A monitoring area information acquisition module is used to acquire N monitoring points, prioritize the N monitoring points, obtain monitoring point sequence information, preset monitoring acquisition frequency, sequentially input theoretical coordinate information and preset error range of target monitoring points according to the monitoring point sequence information, and generate monitoring area information according to the theoretical coordinate information and the preset error range; A monitoring image information acquisition module, which is used to automatically navigate to the monitoring area information with a built-in high-precision navigation module, fix a target icon on the target monitoring point, and use the target icon as a reference to acquire the target monitoring image information of the monitoring area information through a high-precision telescope; A coordinate information confirmation module is used to train and construct a point tracking module, and use the point tracking module to perform target positioning and coordinate extraction on the target monitoring image information to determine the target pixel coordinate information; The point measurement result confirmation module is used to calculate the target coordinates based on the telescope camera parameters and the target pixel coordinate information to obtain the actual space coordinates of the target, and store the actual space coordinates of the target in sequence to generate N spatial point measurement results.

Images