Data acquisition method, system, total station and computer readable storage medium

Abstract

The application provides a data acquisition method and system, a total station and a computer readable storage medium, and relates to the technical field of visual measurement. The method is applied to a data acquisition system, and the data acquisition system comprises a control component, a servo component, an image acquisition component and a measurement component. The method comprises the following steps: through the control component, the servo component is controlled to move to a plurality of acquisition positions according to the zoning information of a measurement area; through the image acquisition component, a zoning image of each acquisition position is acquired; through the measurement component, depth detection is performed based on the zoning image to obtain depth information; and the zoning image and the depth information are used for modeling to generate a target model. In the process of data acquisition, each area is separately acquired and processed through zoning processing, accurate depth information is determined for each different position, the efficiency and accuracy of data acquisition are effectively improved, and the efficiency of modeling is improved.

Description

Technical Field

[0001] This application relates to the field of visual measurement technology, and more specifically, to a data acquisition method, system, total station, and computer-readable storage medium. Background Technology

[0002] Visual SLAM (Simultaneous Localization and Mapping) uses cameras to perceive the environment and can create 3D models of objects. Current visual SLAM technologies include monocular, binocular (multi-view), and RGBD methods.

[0003] Visual SLAM can be applied to surveying instruments such as total stations. Traditional total stations can achieve high-precision spatial point measurement. However, current robotic total stations have slow distance measurement speeds, and the more complex the object's texture, the denser the points need to be scanned, thus increasing the time required. Typically, they can only automatically scan and measure simple, regular objects (such as grain silos and oil depots). Most complex features require manual measurement and mapping of feature points, followed by complex post-processing software to create a model. When facing complex objects or scenes, high-precision measurement and modeling are impossible, or the process is extremely time-consuming. Automated 3D measurement and modeling are not feasible, resulting in low data acquisition efficiency and accuracy for current total stations, leading to inefficient modeling during measurement and failing to meet measurement requirements. Summary of the Invention

[0004] In view of this, the purpose of this application is to provide a data acquisition method, system, total station, and computer-readable storage medium to improve the low efficiency and accuracy of total station data acquisition in the prior art.

[0005] To address the aforementioned problems, in a first aspect, embodiments of this application provide a data acquisition method applied to a data acquisition system, the data acquisition system comprising: a control component, a servo component, an image acquisition component, and a measurement component; the method comprising:

[0006] The control component controls the servo component to move to multiple acquisition positions based on the partition information of the measurement area.

[0007] The image acquisition component acquires a partitioned image at each acquisition location;

[0008] Depth information is obtained by performing depth detection based on the partitioned image using the measurement component.

[0009] The partitioned image and depth information are used for modeling to generate the target model.

[0010] In the above implementation process, during data acquisition, a partitioned processing method can be used to control the servo component to move to multiple acquisition positions corresponding to multiple partitions. This allows for individual image acquisition and depth detection in each region, thereby determining accurate depth information for each different location. The image data and depth information are then combined for appropriate modeling processing. This effectively improves the efficiency of data acquisition and the accuracy of the acquired partitioned images and depth information, thus enhancing modeling efficiency. It eliminates the need for manual acquisition and modification, enabling automated 3D measurement and modeling processing, reducing time and labor costs, and meeting the measurement needs of various scenarios and terrains.

[0011] Optionally, the data acquisition system further includes a data processing component; the step of performing depth detection based on the partitioned image through the measurement component to obtain depth information includes:

[0012] The data processing component is used to identify each of the partitioned images to extract multiple feature points;

[0013] The control component controls the servo component to aim at multiple feature points;

[0014] The measurement component performs depth detection on each feature point when it is pointed at, in order to obtain the depth information.

[0015] In the above implementation process, during depth detection, the data processing component can process the partitioned image to identify and extract multiple feature points in the image. The servo component is then controlled to move and aim at each feature point, allowing the measurement component to perform corresponding depth detection on each feature point during aiming. This enables targeted depth detection of feature points in each region, allowing for separate detection of the depth of different terrains and objects within the region. This improves the accuracy and effectiveness of depth information and is suitable for measuring areas with various complex terrains.

[0016] Optionally, the step of identifying each of the partitioned images through the data processing component to extract multiple feature points includes:

[0017] The data processing component is used to identify the partitioned image in order to obtain the target object;

[0018] The data processing component performs edge extraction on each target object to obtain edge information.

[0019] The data processing component determines the target type of the corresponding target object based on each edge information.

[0020] The data processing component determines multiple feature points based on the edge information and the corresponding target type.

[0021] In the above implementation process, when extracting feature points, multiple target objects in the partitioned image can be identified using image recognition technology in the data processing component. These target objects are then segmented to extract corresponding edge information. Based on this edge information, the target type of the corresponding target object is determined, enabling the identification and segmentation of various objects such as buildings, bridges, and roads within the measurement area, and facilitating the classification of these objects. Furthermore, feature extraction is performed on the target objects by combining edge information and target type, resulting in multiple corresponding feature points. By effectively identifying various types of terrain features within the measurement area, multiple object categories can be classified, and corresponding feature extraction can be performed based on these categories. This improves the effectiveness and accuracy of the extracted feature points, enabling automatic identification, classification, and feature extraction. It also increases the efficiency of feature point acquisition and is suitable for complex terrains with numerous and diverse objects.

[0022] Optionally, the measurement component includes an angle measuring component and a distance measuring component; the step of performing depth detection on each feature point to obtain the depth information by means of the measurement component when each feature point is aimed includes:

[0023] The angle measuring component is used to detect the angle of each feature point to obtain angle information;

[0024] The distance information is obtained by performing distance detection on each feature point using the ranging component.

[0025] The control component performs depth conversion based on the angle information and the distance information to obtain the depth information.

[0026] In the above implementation process, to detect the depth information of each feature point, the angle measurement component and the distance measurement component can be used to measure the angle and distance information corresponding to the feature point, respectively. Then, by combining the angle and distance information, a precise depth-to-relative depth conversion is performed to obtain the corresponding depth information. This allows for the measurement and calculation of the depth information of each feature point from both angle and distance perspectives, improving the accuracy and effectiveness of the depth information.

[0027] Optionally, the method for generating the target model includes:

[0028] The transformation matrix is ​​obtained by weighting the depth information.

[0029] An initial model is obtained by modeling based on the transformation matrix and each of the partitioned images;

[0030] The model objects in the initial model are identified to obtain multiple model attributes;

[0031] Each model object is labeled based on the model attributes to obtain the target model.

[0032] In the above implementation process, during model construction, depth information can be processed accordingly to determine the transformation matrix between the actual depth and the measured depth. Based on the transformation matrix, an initial 3D model is constructed, and model objects are identified and labeled within this initial model to obtain a complete target model. This approach enables the construction of corresponding 3D models based on accurate depth information and partitioned images, effectively improving the modeling efficiency and accuracy of 3D target models without requiring manual processing, thus achieving automated 3D model construction.

[0033] Optionally, the step of controlling the servo component to move to multiple acquisition positions according to the partition information of the measurement area via the control component includes:

[0034] The control component divides the measurement area into partitions according to measurement requirements, thereby obtaining partition information corresponding to multiple sub-regions.

[0035] The control component determines the acquisition location corresponding to each sub-region based on the partition information.

[0036] The control component controls the servo component to move from its current position to each of the acquisition positions.

[0037] In the above implementation process, due to the large area of ​​the measurement region, in order to perform complete and accurate measurements, the control component can divide the measurement region into multiple sub-regions according to the measurement requirements, thereby determining the appropriate acquisition position for each sub-region. The control component sends the position and pose data of the acquisition position to the servo component to control the servo component to move from the current position to the corresponding acquisition position. By partitioning the measurement region, each smaller sub-region can be processed independently, effectively improving the accuracy of image acquisition and depth detection in each sub-region.

[0038] Optionally, the servo component includes a vertical axis and a horizontal axis; the step of acquiring a partition image at each acquisition position through the image acquisition component includes:

[0039] The image acquisition component is driven to move coaxially through the vertical axis and the horizontal axis;

[0040] When the image acquisition component moves to the acquisition position, it acquires an image of the sub-region corresponding to the acquisition position to obtain the partitioned image.

[0041] In the above implementation process, the servo component can rotate vertically and horizontally along the vertical and horizontal axes, enabling it to accurately move to the corresponding acquisition position. Furthermore, the servo component can drive the image acquisition component to move coaxially, allowing the image acquisition component to move synchronously with the servo component and to move to various acquisition positions to acquire images of the corresponding sub-regions at each position, obtaining the corresponding partitioned images. Driving the image acquisition component through coaxial motion improves the accuracy of positioning during movement and the efficiency of image acquisition.

[0042] Secondly, embodiments of this application also provide a data acquisition system, the system comprising: a control component, a servo component, an image acquisition component, and a measurement component;

[0043] The control component is used to control the servo component to move to multiple acquisition positions according to the partition information of the measurement area;

[0044] The image acquisition component is used to acquire partitioned images at each acquisition location;

[0045] The measurement component is used to perform depth detection based on the partitioned image to obtain depth information;

[0046] The partitioned image and depth information are used for modeling to generate the target model.

[0047] In the above implementation process, in the data acquisition system, the control component controls the servo component to move to multiple acquisition positions corresponding to multiple partitions in a partitioned processing manner. The image acquisition component and the measurement component perform separate image acquisition and depth detection for each region, so as to combine the image data and depth information for corresponding modeling processing.

[0048] Thirdly, embodiments of this application also provide a total station, which includes the data acquisition system described above.

[0049] Fourthly, embodiments of this application also provide a computer-readable storage medium storing computer program instructions, which, when read and executed by a processor, perform steps in any of the above-described implementations of the data acquisition method.

[0050] In summary, the embodiments of this application provide a data acquisition method, system, total station, and computer-readable storage medium. By employing a partitioned processing approach, each area is individually acquired and processed. Accurate image data and depth information are determined for each different location, and the image data and depth information are combined for corresponding modeling processing. This effectively improves the efficiency of data acquisition and the accuracy of the acquired partitioned images and depth information, thereby improving modeling efficiency. No manual acquisition or modification is required, enabling automated 3D measurement and modeling processing, reducing time and labor costs, and meeting the measurement needs of various scenarios and terrains. Attached Figure Description

[0051] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0052] Figure 1 This is a schematic diagram of the structure of a data acquisition system provided in an embodiment of this application;

[0053] Figure 2 A detailed structural diagram of a data acquisition system provided in an embodiment of this application;

[0054] Figure 3 A flowchart illustrating a data acquisition method provided in an embodiment of this application;

[0055] Figure 4 A detailed flowchart of step S400 provided for an embodiment of this application;

[0056] Figure 5 A detailed flowchart of step S410 provided for an embodiment of this application;

[0057] Figure 6 A detailed flowchart of step S430 provided for an embodiment of this application;

[0058] Figure 7 A detailed flowchart illustrating another data acquisition method provided in this application embodiment;

[0059] Figure 8 A detailed flowchart of step S200 provided for an embodiment of this application;

[0060] Figure 9 This is a detailed flowchart of step S300 provided in an embodiment of this application.

[0061] Icons: 100 - Data acquisition system; 110 - Control component; 120 - Servo component; 121 - Vertical axis; 122 - Horizontal axis; 130 - Image acquisition component; 140 - Measurement component; 141 - Angle measurement component; 142 - Distance measurement component; 150 - Data processing component. Detailed Implementation

[0062] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of the embodiments of this application.

[0063] Current visual SLAM technologies include monocular, binocular (multi-view), and RGBD methods. Monocular solutions cannot perceive distance and require using triangulation to calculate the camera's position and perceive 3D spatial information. They can only perceive relative depth information and cannot simultaneously magnify the object and its distance. Binocular (multi-view) systems can calculate depth directly without motion and can perceive the object and its own trajectory for 3D modeling. Currently, binocular systems are widely used, typically employing a "dual camera + IMU (Inertial Measurement Unit)" multi-sensor time synchronization and multi-source depth fusion calculation to correct errors, meeting the needs of SLAM research, intelligent robots, drone obstacle avoidance, and indoor and outdoor navigation and positioning. RGBD cameras are a new type of camera. These sensors have built-in distance information and can directly measure the distance of each pixel. Color modeling can be achieved through dedicated software, but the distance measurement range is usually only a few meters, limiting its application scope.

[0064] The inventors of this application have discovered that when SLAM technology is applied to surveying instruments such as total stations, traditional total stations can achieve high-precision spatial point measurements. However, current robotic total stations have slow distance measurement speeds; for example, the average measurement time for each point is over 2 seconds. The more complex the object's texture, the denser the points need to be scanned, thus increasing the time required. Typically, only simple, regular objects (such as grain silos and oil depots) can be automatically scanned and measured. Most complex features require manual measurement and mapping of feature points, followed by complex post-processing software for modeling. When facing complex objects or scenes, high-precision measurement and modeling are impossible, or the measurement and modeling processes are extremely time-consuming. Automated 3D measurement and modeling are not possible, resulting in low efficiency and accuracy of current total station data acquisition, leading to inefficient modeling during measurement and failing to meet measurement requirements.

[0065] Therefore, in order to solve the above problems, this application provides a data acquisition method that is applied to a data acquisition system. The data acquisition system can be set in a total station or other measuring equipment, and can efficiently and accurately measure images and corresponding depth information to achieve automated 3D modeling based on the acquired image data and depth information.

[0066] Please see Figure 1 , Figure 1 This is a schematic diagram of the structure of a data acquisition system provided in an embodiment of this application. The data acquisition system 100 may include: a control component 110, a servo component 120, an image acquisition component 130, and a measurement component 140.

[0067] The control component 110 is used to control the servo component 120 to move to multiple acquisition positions according to the partition information of the measurement area;

[0068] The image acquisition component 130 is used to acquire partitioned images at each acquisition location;

[0069] Measurement component 140 is used to perform depth detection based on partitioned images to obtain depth information;

[0070] In this process, the partitioned image and depth information are used for modeling to generate the target model.

[0071] Optionally, such as Figure 1 As shown, in the data acquisition system 100, the control component 110 is connected to the servo component 120 to control the movement of the servo component 120; the servo component 120 is connected to the image acquisition component 130 and the measurement component 140 so that the image acquisition component 130 and the measurement component 140 can perform data acquisition and measurement based on the movement of the servo component 120. It should be noted that... Figure 1 Only one preferred or reference connection method is shown in this application; other connection methods or arrangements are also within the scope of this application.

[0072] Optionally, the data acquisition system 100 can connect to external electronic devices via communication components in surveying equipment such as a total station. These electronic devices can be servers, personal computers (PCs), tablets, smartphones, personal digital assistants (PDAs), or other devices with logical computing capabilities. After the data acquisition system 100 obtains the zoning image and depth information, real-time modeling processing can be performed directly in the total station during the data measurement process to generate a target model corresponding to the measurement area. This target model is then sent to the electronic device for subsequent processing. Alternatively, the zoning image and depth information can be sent to the connected electronic device for real-time modeling processing during the measurement process, or the electronic device can perform modeling processing after the data measurement is completed to generate a target model corresponding to the measurement area.

[0073] Optionally, the control component 110 can be a processor of multiple types. A processor can be an integrated circuit chip with signal processing capabilities. The processor can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it can also be a digital signal processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor can be a microprocessor or any conventional processor.

[0074] Optionally, please refer to Figure 2 , Figure 2 This is a detailed structural diagram of a data acquisition system provided in an embodiment of this application.

[0075] In an optional implementation, the data acquisition system 100 may further include a data processing component 150, which identifies each partitioned image to extract multiple feature points; a control component 110 controls a servo component 120 to aim at the multiple feature points; and a measurement component 140 performs depth detection on each feature point when it is aimed at it to obtain depth information.

[0076] In an optional implementation, the data processing component 150 identifies the partitioned image to obtain the target object; the data processing component 150 performs edge extraction on each target object to obtain edge information; the data processing component 150 determines the target type of the corresponding target object based on each edge information; and the data processing component 150 determines multiple feature points based on the edge information and the corresponding target type.

[0077] In an optional implementation, the measurement component 140 may include an angle measuring component 141 and a distance measuring component 142; the angle measuring component 141 performs angle detection on each feature point to obtain angle information; the distance measuring component 142 performs distance detection on each feature point to obtain distance information; and the control component 110 performs depth conversion based on the angle information and distance information to obtain depth information.

[0078] In an optional implementation, the method for generating the target model may include: performing weighted processing based on depth information to obtain a transformation matrix; modeling based on the transformation matrix and each partitioned image to obtain an initial model; identifying model objects in the initial model to obtain multiple model attributes; and labeling each model object based on the model attributes to obtain the target model.

[0079] In an optional implementation, the control component 110 divides the measurement area into partitions according to measurement requirements to obtain partition information corresponding to multiple sub-regions; the control component 110 determines the acquisition position corresponding to each sub-region based on the partition information; and the control component 110 controls the servo component 120 to move from the current position to each acquisition position.

[0080] In an optional embodiment, the servo component 120 includes a vertical axis 121 and a horizontal axis 122; the vertical axis 121 and the horizontal axis 122 drive the image acquisition component 130 to move coaxially; when the image acquisition component 130 moves to the acquisition position, it acquires an image of the sub-region corresponding to the acquisition position to obtain a partitioned image.

[0081] Optionally, such as Figure 2As shown, in the data acquisition system 100, the angle measuring component 141 can be divided into a horizontal angle measuring component and a vertical angle measuring component (not shown). The angle measuring component 141 can be mounted on the servo component 120. The control component 110 is connected to the vertical axis 121 for vertical movement and is also connected to the horizontal axis 122 for horizontal movement. The distance measuring component 142 can be located in the same area as the image acquisition component 130. The angle measuring component 141, distance measuring component 142, and image acquisition component 130 are all connected to the data processing component 150 to send the acquired images and data to the data processing component 150 for processing. It should be noted that... Figure 2 Only one preferred or reference connection method is shown in this application; other connection methods or arrangements are also within the scope of this application.

[0082] The data acquisition system in this embodiment can be used to execute the various steps in the data acquisition methods provided in the embodiments of this application. The implementation process of the data acquisition methods is described in detail below through several embodiments.

[0083] Optionally, please refer to Figure 3 , Figure 3 This is a flowchart illustrating a data acquisition method provided in an embodiment of this application. The method may include steps S200-S400.

[0084] In step S200, the control component controls the servo component to move to multiple acquisition positions based on the partition information of the measurement area.

[0085] Because some measurement areas are large or have complex terrain and buildings, the control component can partition the measurement area according to the actual conditions to obtain corresponding partition information. This partition information can include the boundaries, area, center point, and other information of each sub-region. The control component can also determine the appropriate acquisition location for each sub-region based on the partition information and send the location, pose, and other data of the acquisition location to the servo component, enabling the servo component to move to the corresponding acquisition location.

[0086] Step S300: The image acquisition component acquires the partition image of each acquisition location.

[0087] The image acquisition component can move coaxially with the servo component and can also move to the corresponding acquisition position. When the servo component moves to the acquisition position, it can stay for a preset time period so that the image acquisition component can acquire images of the area corresponding to the acquisition position and obtain the corresponding partitioned image.

[0088] Optionally, the image acquisition component can be various types of devices with image acquisition capabilities, such as various types of cameras.

[0089] Step S400: Depth information is obtained by performing depth detection based on the partitioned image using the measurement component.

[0090] The measurement component can perform depth detection on the acquired partitioned image to detect the depth information contained in the partitioned image.

[0091] Optionally, the acquired partitioned images and depth information are used to build a 3D model of the measurement area, i.e., to generate the target model. AI deep learning technology can be used for modeling. Current AI deep learning technology can perform 3D reconstruction from single or multiple images, predicting depth spatial information based on multi-view matching and image-level pixel texture. This allows the acquisition of 3D information of ground features. However, the depth values ​​obtained by reconstructing ground features such as buildings entirely from images are the relative distances between features. The constructed 3D model only reflects the spatial relationships between ground features, resulting in low modeling accuracy when using currently measured data. In this application, the depth information obtained from data acquisition is high-precision spatial point information. Therefore, a high-precision 3D model can be constructed based on the acquired partitioned images and depth information, with a typical accuracy reaching 2mm / 500m.

[0092] exist Figure 3 The embodiments shown can effectively improve the efficiency of data acquisition and the accuracy of the acquired partitioned images and depth information, thereby improving the efficiency of modeling. No manual acquisition and modification is required, and automated 3D measurement and modeling processing can be achieved, reducing time and labor costs and meeting the measurement needs of various scenarios and terrains.

[0093] Optionally, please refer to Figure 4 , Figure 4 This is a detailed flowchart of step S400 provided in an embodiment of the present application. Step S400 may also include steps S410-S430.

[0094] In step S410, each partitioned image is identified by the data processing component to extract multiple feature points.

[0095] Since the partitioned images can include a variety of objects in the measurement area, such as trees, roads, bridges, utility poles, iron towers, road slopes, conventional buildings, door frames, window frames, corners, roof corners, walls, fences, etc., the data processing component can identify each partitioned image to extract multiple feature points corresponding to the objects within it.

[0096] Optionally, the data processing component can be configured with various image recognition algorithms, such as AI intelligent recognition algorithms, to identify objects present in the partitioned image.

[0097] In step S420, the control component controls the servo component to aim at multiple feature points.

[0098] Since the feature points are located at different positions, the control component can determine the aiming position of the servo component when aiming at the feature point based on the relative position of the feature point and the servo component. This allows the servo component to move to the aiming position to measure the feature point, effectively improving the accuracy of the feature point measurement.

[0099] In step S430, depth detection is performed on each feature point when it is aimed at, using the measurement component to obtain depth information.

[0100] The measuring component can move synchronously with the servo component. When the servo component moves to the corresponding aiming position, it can also stay for a preset time so that the measuring component can perform depth detection on each feature point when each feature point is aimed at, so as to obtain high-precision spatial point information as depth information, which effectively improves the accuracy and effectiveness of the acquired depth information.

[0101] Optionally, the depth information can be in the form of a depth map or other data formats.

[0102] exist Figure 4 In the illustrated embodiment, targeted depth detection can be performed on feature points in each region, thereby enabling separate detection of the depth of different terrains and objects within the region, improving the accuracy and effectiveness of depth information, and making it suitable for measuring regions with various complex terrains.

[0103] Optionally, please refer to Figure 5 , Figure 5 This is a detailed flowchart of step S410 provided in an embodiment of the present application. Step S410 may also include steps S411-S414.

[0104] Step S411: The partitioned image is identified by the data processing component to obtain the target object.

[0105] In the process of extracting feature points, multiple objects in the partitioned image, such as trees, roads, bridges, utility poles, iron towers, road slopes, conventional buildings, door frames, window frames, corners, roof corners, walls, fences, etc., can be identified according to the preset image recognition algorithm in the data processing component to obtain the corresponding multiple target objects.

[0106] Step S412: Edge extraction is performed on each target object using the data processing component to obtain edge information.

[0107] Among them, image recognition technology can be used to extract and cut the edges of each identified target object to obtain the corresponding edge information. The edge information can include various information such as the outline and shape of the target object.

[0108] Step S413: Using the data processing component, determine the target type of the corresponding target object based on each edge information.

[0109] Among them, image recognition technology, such as AI recognition algorithms, can accurately identify and segment objects of multiple categories and in multiple scenes with the support of a large amount of data in the field of recognition and segmentation. Therefore, in complex scenes, the use of deep learning technology can efficiently and accurately classify the object type of each target object based on the extracted edge information, and obtain the corresponding target type. For example, street lights and utility poles are classified as road elements, buildings and iron towers are classified as buildings, and door frames and window frames are classified as building elements.

[0110] Step S414: Using the data processing component, determine multiple feature points based on edge information and the corresponding target type.

[0111] Among them, edge information and target type can be combined to perform feature extraction, high-dimensional data calculation, feature matching and other processing on the target object, so as to extract the corresponding feature points on the edge contour of the target object, and still effectively identify the feature points of the target object in a specific area in a complex environment.

[0112] exist Figure 5 In the illustrated embodiment, by effectively identifying various types of land features in the measurement area, it is possible to classify multiple objects into categories and extract corresponding features based on their categories. This improves the effectiveness and accuracy of the extracted feature points, enables automatic identification, classification, and feature extraction, and also improves the efficiency of feature point acquisition. It is suitable for complex terrains with a large number of objects and a wide variety of object types.

[0113] Optionally, please refer to Figure 6 , Figure 6 This is a detailed flowchart of step S430 provided in an embodiment of the present application. Step S430 may also include steps S431-S433.

[0114] Step S431: Use the angle measuring component to detect the angle of each feature point and obtain the angle information.

[0115] The measurement component may include an angle measuring component and a distance measuring component. The angle measuring component can detect the angle between the feature point and the test position to obtain the corresponding angle information.

[0116] Optionally, the angle measuring component may also include a horizontal angle measuring unit and a vertical angle measuring unit to measure the horizontal angle and the vertical angle respectively, and obtain the corresponding angle information.

[0117] Step S432: Use the ranging component to perform distance detection on each feature point to obtain distance information.

[0118] Among them, the ranging component can detect the distance between the feature point and the test position to obtain the corresponding distance information.

[0119] In step S433, depth information is obtained by performing depth conversion based on angle and distance information through the control component.

[0120] In this process, corresponding angle and distance information can be matched for each feature point to achieve the conversion between precise depth and relative depth, thereby obtaining the corresponding depth information.

[0121] exist Figure 6 In the illustrated embodiment, the depth information of each feature point can be measured and calculated from both angle and distance perspectives, thereby improving the accuracy and effectiveness of the depth information.

[0122] Optionally, please refer to Figure 7 , Figure 7 This is a detailed flowchart illustrating another data acquisition method provided in an embodiment of this application. The method may further include steps S510-S540.

[0123] Step S510: Perform weighted processing based on depth information to obtain the transformation matrix.

[0124] The relationship between the actual depth and the relative depth can be calculated using a weighted fusion method, thereby obtaining the transformation matrix between the two.

[0125] Step S520: Model the image based on the transformation matrix and each partition image to obtain the initial model.

[0126] During modeling, since there are multiple different partitioned images, each with its own unique characteristics, the modeling process can begin by using deep learning techniques and a transformation matrix to perform AI-powered 3D modeling based on each partitioned image and its corresponding depth information. This process generates multiple sub-models, which are then registered across all partitioned images. Finally, these sub-models are stitched together to achieve a larger-scale 3D model and obtain a complete initial model of the entire measurement area.

[0127] Step S530: Identify the model objects in the initial model to obtain multiple model attributes.

[0128] After obtaining the overall initial model, image recognition technology can be used to identify the attributes of all model objects in the model, resulting in multiple model attributes.

[0129] Optionally, model attributes may include multiple related parameters such as the model object's outline, size, type, and number.

[0130] Step S540: Annotate each model object based on the model attributes to obtain the target model.

[0131] In the data processing component, when extracting feature points based on image recognition technology, it can automatically encode the extracted target object and target type to generate corresponding encoding information. This enables the correspondence between the target object and the model object. Based on the encoding information and model attributes, each model object is automatically labeled to obtain a complete target model, which can then be used to interpret the attributes of each object in the constructed model.

[0132] Optionally, the annotation process can also define the required annotation range based on needs, so that all objects within the defined range can be annotated.

[0133] Optionally, the model attributes may also include color information of the target object, and the corresponding model object may be colored based on the color information according to the correspondence between the target object and the model object, so as to achieve the automatic coloring effect of the model.

[0134] exist Figure 7 In the illustrated embodiment, a corresponding 3D model can be constructed based on the partitioned image using accurate depth information, which effectively improves the modeling efficiency and accuracy of the 3D target model, and achieves automated 3D model construction without manual processing.

[0135] Optionally, please refer to Figure 8 , Figure 8 This is a detailed flowchart of step S200 provided in an embodiment of the present application. Step S200 may also include steps S210-S230.

[0136] Step S210: The measurement area is divided into partitions according to the measurement requirements by the control component to obtain partition information corresponding to multiple sub-regions.

[0137] Since the measurement area is relatively large, in order to measure the area completely and accurately, the control component can divide the measurement area according to the measurement requirements to obtain multiple sub-regions and corresponding partition information. The measurement requirements can include various information such as the number of sub-regions and the division rules, while the partition information can include various information such as the boundaries, area, and center point of each sub-region after partitioning.

[0138] Step S220: Using the control component, determine the collection location corresponding to each sub-region based on the partition information.

[0139] Based on various information such as the boundary, area, and center point of each sub-region in the partition information, the optimal acquisition position for the acquisition viewpoint in each sub-region can be determined.

[0140] Step S230: Control the servo component to move from the current position to each acquisition position via the control component.

[0141] The control component sends the position and pose data of the acquisition location to the servo component to control the servo component to move from the current position to the corresponding acquisition position.

[0142] exist Figure 8 In the illustrated embodiment, by partitioning the measurement area, each smaller sub-region can be processed separately, effectively improving the accuracy of image acquisition and depth detection in each sub-region.

[0143] Optionally, please refer to Figure 9 , Figure 9 This is a detailed flowchart of step S300 provided in an embodiment of the present application. Step S300 may also include steps S310-S320.

[0144] In step S310, the image acquisition component is driven to move coaxially through the vertical and horizontal axes.

[0145] The servo component can rotate vertically and horizontally via its vertical and horizontal axes, enabling it to accurately move to the corresponding acquisition position. Furthermore, the servo component can drive the image acquisition component to move coaxially, allowing the image acquisition component to move synchronously with the servo component and to different acquisition positions for image acquisition.

[0146] Step S320: Using the image acquisition component, when the image acquisition component moves to the acquisition position, the sub-region corresponding to the acquisition position is acquired to obtain a partitioned image.

[0147] The image acquisition component can detect its own position. When it detects that it has moved to the corresponding acquisition position and no corresponding image has been acquired at that acquisition position, it can acquire images of the corresponding sub-region at the acquisition position to obtain one or more partitioned images.

[0148] exist Figure 9 In the illustrated embodiment, the image acquisition component is driven to move by coaxial motion, which improves the accuracy of positioning during movement and the efficiency of image acquisition.

[0149] This application also provides a computer-readable storage medium storing computer program instructions. When the computer program instructions are read and executed by a processor, they perform the steps of any of the data acquisition methods provided in this embodiment.

[0150] In the several embodiments provided in this application, it should be understood that the disclosed device can also be implemented in other ways. The system embodiments described above are merely illustrative; for example, the block diagrams in the accompanying drawings show the architecture, functionality, and operation of possible implementations of the device according to various embodiments of this application. In this regard, each block in the block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagram, and combinations of block diagrams, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0151] In addition, the functional modules in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0152] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0153] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application. It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0154] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

[0155] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes said element.

Claims

1. A data acquisition method, characterized in that, The method is applied to a data acquisition system, which includes a control component, a servo component, an image acquisition component, and a measurement component; the method includes: The control component controls the servo component to move to multiple acquisition positions based on the partition information of the measurement area. The image acquisition component acquires a partitioned image at each acquisition location; Depth information is obtained by performing depth detection based on the partitioned image using the measurement component. The partitioned image and depth information are used for modeling to generate the target model; The data acquisition system further includes a data processing component; the step of performing depth detection based on the partitioned images through the measurement component to obtain depth information includes: identifying each partitioned image through the data processing component to extract multiple feature points; controlling the servo component to aim at the multiple feature points through the control component; and performing depth detection on each feature point when each feature point is aimed at through the measurement component to obtain the depth information. The step of identifying each of the partitioned images using the data processing component to extract multiple feature points includes: identifying the partitioned images using the data processing component based on a set AI intelligent recognition algorithm to obtain target objects; performing edge extraction on each target object using the data processing component to obtain edge information; determining the target type of the corresponding target object using the data processing component based on each edge information; and determining multiple feature points using the data processing component based on the edge information and the corresponding target type.

2. The method according to claim 1, characterized in that, in, The measurement component includes an angle measuring component and a distance measuring component; the measurement component performs depth detection on each feature point when each feature point is aimed at, to obtain the depth information, including: The angle measuring component is used to detect the angle of each feature point to obtain angle information; The distance information is obtained by performing distance detection on each feature point using the ranging component. The control component performs depth conversion based on the angle information and the distance information to obtain the depth information.

3. The method according to any one of claims 1-2, characterized in that, in, The generation method of the target model includes: The transformation matrix is ​​obtained by weighting the depth information. An initial model is obtained by modeling based on the transformation matrix and each of the partitioned images; The model objects in the initial model are identified to obtain multiple model attributes; Each model object is labeled based on the model attributes to obtain the target model.

4. The method according to any one of claims 1-2, characterized in that, The step of controlling the servo component to move to multiple acquisition positions based on the partition information of the measurement area, through the control component, includes: The control component divides the measurement area into partitions according to measurement requirements, thereby obtaining partition information corresponding to multiple sub-regions. The control component determines the acquisition location corresponding to each sub-region based on the partition information. The control component controls the servo component to move from its current position to each of the acquisition positions.

5. The method according to any one of claims 1-2, characterized in that, in, The servo component includes a vertical axis and a horizontal axis; the acquisition of partitioned images at each acquisition position via the image acquisition component includes: The image acquisition component is driven to move coaxially through the vertical axis and the horizontal axis; When the image acquisition component moves to the acquisition position, it acquires an image of the sub-region corresponding to the acquisition position to obtain the partitioned image.

6. A data acquisition system, characterized in that, The system includes: a control component, a servo component, an image acquisition component, and a measurement component; The control component is used to control the servo component to move to multiple acquisition positions according to the partition information of the measurement area; The image acquisition component is used to acquire partitioned images at each acquisition location; The measurement component is used to perform depth detection based on the partitioned image to obtain depth information; The partitioned image and depth information are used for modeling to generate the target model; The data acquisition system further includes a data processing component; the data processing component is used to identify each of the partitioned images to extract multiple feature points; the control component is used to control the servo component to aim at the multiple feature points; the measurement component is used to perform depth detection on each feature point when it is aimed at it to obtain the depth information. The data processing component is specifically used for: identifying the partitioned image based on a set AI intelligent recognition algorithm to obtain target objects; extracting edges from each target object to obtain edge information; determining the target type of the corresponding target object based on each edge information; and determining multiple feature points based on the edge information and the corresponding target type.

7. A total station, characterized in that, The total station includes the data acquisition system described in claim 6.

8. A computer-readable storage medium, characterized in that, The readable storage medium stores computer program instructions, which, when executed by a processor, perform the steps of the method according to any one of claims 1-5.

Images