A concrete vibrating operation real-time guidance method, device, equipment and storage medium

Abstract

This invention relates to the field of machine vision technology, providing a method, device, equipment, and storage medium for real-time guidance of concrete vibration operations. By generating a digital map and monitoring grid with physical scale relationships, it dynamically detects changes in the concrete surface during construction and automatically updates the digital map and monitoring grid. Through instance segmentation models, it identifies the vibration personnel and the vibrator to determine whether the vibrator is in an effective vibration state. It extracts equivalent vibration points and locates them on the digital map. By statistically analyzing the vibration coverage within the monitoring grid and comparing it with preset construction standards, it automatically identifies under-vibration or over-vibration areas, thereby generating spatially directional guidance information or early warning information. This invention achieves adaptive perception and precise operation guidance of dynamic construction surfaces using a low-cost monocular vision solution.

Description

Technical Field

[0001] This invention relates to the field of machine vision technology, and in particular to a method, apparatus, equipment and storage medium for real-time guidance of concrete vibration operations. Background Technology

[0002] In concrete construction, vibration is a crucial step in ensuring concrete density and eliminating internal air bubbles and honeycomb-like pitting. Currently, this step relies heavily on the experience of construction workers, which has the following significant drawbacks:

[0003] (1) Vibration is highly arbitrary and lacks quantitative standards: Construction workers control the vibration points and duration based on experience, which can easily lead to missed vibration, under-vibration or over-vibration, affecting the final quality of concrete.

[0004] (2) Incomplete coverage and lack of process monitoring: It is impossible to record the vibrated and unvibrated areas in real time and objectively, making it difficult to ensure full coverage of the radius of influence of the vibrator, and making it difficult to trace and hold accountable afterwards;

[0005] (3) Lack of in-process guidance: Most existing technical solutions are based on post-event analysis and cannot provide real-time and accurate guidance to operators during construction, nor can they correct incorrect vibration behavior in a timely manner.

[0006] Furthermore, most existing vision-based vibration control solutions employ binocular stereo vision systems. While binocular vision solutions can acquire depth information, they suffer from problems such as complex system structure, cumbersome calibration, high hardware costs, and large computational loads, making them difficult to promote on a large scale in construction sites. On the other hand, traditional monocular vision solutions lack effective spatial mapping and quantitative management methods, making it difficult to convert two-dimensional image information into three-dimensional spatial instructions that directly guide construction personnel.

[0007] In view of this, it is necessary to propose a method, device, equipment and storage medium for real-time guidance of concrete vibration operations to solve or at least alleviate the above-mentioned defects. Summary of the Invention

[0008] The main objective of this invention is to provide a method, device, equipment, and storage medium for real-time guidance of concrete vibration operations, in order to solve the technical problem that existing concrete vibration construction processes rely on manual experience, resulting in a lack of dynamic benchmarks and an inability to provide real-time and accurate guidance.

[0009] To achieve the above objectives, the present invention provides a real-time guidance method for concrete vibration operations, comprising the following steps:

[0010] S1, acquire a construction site image containing multiple pre-deployed fixed spatial reference points, as well as the world coordinates and pixel coordinates of each fixed spatial reference point; calculate spatial transformation parameters based on the pixel coordinates and the world coordinates;

[0011] S2, based on the spatial transformation parameters, the vibratory compaction area in the construction site image is mapped to generate a digital map with a real physical scale relationship; the effective range of the vibrator is obtained, and a monitoring grid is divided on the digital map according to the effective range;

[0012] S3, acquire the on-site video stream of the construction process, perform frame extraction processing on the on-site video stream according to a preset period and identify the current concrete surface, and then determine whether the current concrete surface meets the preset change conditions; if so, recalculate the spatial transformation parameters based on the fixed spatial reference point and update the digital map and monitoring grid.

[0013] S4, determine whether the vibrator in the image is in an effective vibration operation state; if so, determine the equivalent vibration point in the image, and convert the pixel coordinates of the equivalent vibration point to the world coordinates on the digital map according to the spatial transformation parameters.

[0014] S5, map the converted equivalent vibration points to the updated monitoring grid, and count the vibration coverage of each monitoring grid; compare the vibration coverage with the preset construction standard to identify under-vibrated or over-vibrated grid units; wherein, generate directional guidance information pointing to the under-vibrated grid unit for the under-vibrated grid unit; generate early warning information for the over-vibrated grid unit.

[0015] Preferably, the step S2 of obtaining the effective range of the vibrator and dividing the monitoring grid on the digital map according to the effective range includes the following steps:

[0016] S21, obtain the effective radius of action of the vibrator, and determine the physical side length of the monitoring grid unit based on the effective radius of action;

[0017] S22, calculate the scale parameter of the digital map, and calculate the pixel size corresponding to each monitoring grid according to the effective radius of action and the scale parameter;

[0018] S23, draw grid lines on the digital map to form the monitoring grid according to the pixel size.

[0019] Preferably, step S3 includes the following steps:

[0020] S31, acquire the on-site video stream of the construction process, extract frames from the on-site video stream according to a preset time interval, process the extracted images using a pre-trained first instance segmentation model to obtain the segmentation mask of the current concrete slab surface, and then extract the outline of the concrete slab surface from the segmentation mask.

[0021] S32, calculate the outline area based on the concrete sump surface outline, calculate the intersection point between the concrete sump surface outline and the outline of the marker reference object associated with the fixed spatial reference point, and then obtain the pixel coordinates of the intersection point;

[0022] S33, calculate the rate of change of the contour area compared to the previously updated contour area, and calculate the actual physical displacement of the pixel coordinates of the intersection point after being transformed by the current spatial transformation parameters.

[0023] S34, determine whether the area change rate is greater than a first preset threshold, and determine whether the actual physical displacement of at least one of the intersection points is greater than a second preset threshold.

[0024] S35, when both of the following conditions are met simultaneously: the area change rate is greater than the first preset threshold and the actual physical displacement with at least one of the intersection points is greater than the second preset threshold, the current concrete surface is determined to meet the preset change conditions.

[0025] S36. Based on the current concrete surface outline and the fixed spatial reference point, recalculate the spatial transformation parameters and update the digital map and monitoring grid based on the recalculated spatial transformation parameters.

[0026] Preferably, step S4, determining whether the vibrator in the image is in an effective vibration operation state, includes the following steps:

[0027] The pre-trained second instance segmentation model is used to identify the segmentation mask of the vibrating personnel and the segmentation mask of the vibrating rod in each frame of the image.

[0028] The cross-union ratio (CUNR) of the vibration personnel segmentation mask and the vibration rod segmentation mask in the identification results is calculated. When the CUNR is greater than a set threshold, the vibration rod detection is determined to be correct; when the CUNR is less than or equal to the set threshold, the vibration rod detection result is filtered.

[0029] Perform a geometric intersection operation between the vibrator segmentation mask and the segmentation mask of the current concrete placement surface. If there is an intersection between the vibrator segmentation mask and the segmentation mask of the current concrete placement surface, mark the current operation state as a valid vibration operation state; if there is no intersection between the vibrator segmentation mask and the segmentation mask of the current concrete placement surface, mark the current operation state as an invalid vibration operation state.

[0030] Preferably, determining the equivalent vibration point in the image in step S4 includes the following steps:

[0031] When the current operation is in the effective vibration operation state, extract the contour points of the vibratory rod dividing the mask, and select the point at the bottom of the vibratory rod as the equivalent vibration point based on the contour points.

[0032] Preferably, step S5, which maps the converted equivalent vibration points to the updated monitoring grid and calculates the vibration coverage of each monitoring grid, includes the following steps:

[0033] Using the converted equivalent vibration point as the center and the effective radius of the vibrator as the radius, a circular influence area is determined on the digital map, and the cumulative number of vibrations for each monitoring grid that falls into the circular influence area is increased by one.

[0034] Based on the final cumulative number of vibrations for each monitoring grid, the cumulative vibration duration for each monitoring grid is calculated using a preset time conversion relationship, and the cumulative vibration duration is used as the vibration coverage.

[0035] Preferably, the preset time conversion relationship is: cumulative vibration duration = cumulative vibration times × (K / F); where K is the number of original video frames between two adjacent frames extracted from the on-site video stream, and F is the number of frames per second of the on-site video stream.

[0036] The present invention also provides a real-time guidance device for concrete vibration operation, used to perform the real-time guidance method for concrete vibration operation as described above, comprising:

[0037] The transformation parameter determination unit is used to acquire a construction site image containing multiple pre-laid fixed spatial reference points, as well as the world coordinates and pixel coordinates of each fixed spatial reference point; and to calculate spatial transformation parameters based on the pixel coordinates and the world coordinates.

[0038] The map grid unit is used to map the vibratory compaction area in the construction site image to generate a digital map with a real physical scale relationship according to the spatial transformation parameters; to obtain the effective range of the vibratory rod, and to divide the monitoring grid on the digital map according to the effective range;

[0039] The dynamic update unit is used to acquire on-site video streams during the construction process, perform frame extraction processing on the on-site video streams according to a preset period, identify the current concrete surface, and then determine whether the current concrete surface meets the preset change conditions; if so, the spatial transformation parameters are recalculated based on the fixed spatial reference point, and the digital map and monitoring grid are updated.

[0040] The vibration judgment unit is used to determine whether the vibrator in the image is in an effective vibration operation state; if so, it determines the equivalent vibration point in the image and transforms the pixel coordinates of the equivalent vibration point to the world coordinates on the digital map according to the spatial transformation parameters.

[0041] The guidance and decision-making unit is used to map the converted equivalent vibration points to the updated monitoring grid and to count the vibration coverage of each monitoring grid; based on the vibration coverage, it compares with the preset construction standard to identify under-vibrated or over-vibrated grid units; wherein, for under-vibrated grid units, it generates directional guidance information pointing to the under-vibrated grid units; and for over-vibrated grid units, it generates early warning information.

[0042] The present invention also provides a real-time guidance device for concrete vibration operation, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the real-time guidance method for concrete vibration operation as described above.

[0043] The present invention also provides a storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described real-time guidance method for concrete vibration operation.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] This invention provides a method, device, equipment, and storage medium for real-time guidance of concrete vibration operations. Based on a dynamic visual grid segmentation scheme using marker references, this application solves the industry problem of visual reference failure caused by the rise of the concrete pouring surface. The monitoring grid automatically updates as the concrete pouring surface grows, ensuring synchronization between the digital map, the monitoring grid, and the concrete pouring surface. This guarantees continuous accuracy of detection and guidance throughout the entire vibration cycle, providing a unique and reliable spatiotemporal reference for precise global control. Through instance segmentation models and conditional judgments, it can automatically and in real-time accurately identify whether the vibrator is in a valid operational state from complex video streams, reducing interference from invalid data. Furthermore, this invention innovatively adopts a monocular vision combined with visual grid segmentation technology, replacing the expensive binocular vision system. Image acquisition can be performed using a monocular camera, resulting in low cost and simple deployment, significantly lowering the barrier to technology promotion. Attached Figure Description

[0046] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0047] Figure 1 This is a schematic flowchart of one embodiment of the present invention;

[0048] Figure 2 This is a schematic diagram illustrating an example of data annotation required for the second instance separation model in one embodiment of the present invention;

[0049] Figure 3 This is a schematic diagram illustrating an example of data annotation required for the first instance separation model in one embodiment of the present invention;

[0050] Figure 4 The purple box in one embodiment of the present invention is a schematic diagram of a reference marker selected on-site.

[0051] Figure 5 This is a schematic diagram of the concrete slab surface grid division in one embodiment of the present invention;

[0052] Figure 6 This is a schematic diagram showing the relationship between the concrete surface after pouring and the reference marker in one embodiment of the present invention;

[0053] Figure 7 This is a schematic diagram of the recalculation of the monitoring grid after concrete pouring in one embodiment of the present invention;

[0054] Figure 8 This is a schematic diagram of the equivalent vibration point in one embodiment of the present invention;

[0055] Figure 9 This is a schematic diagram of the vibration mass thermal curve in one embodiment of the present invention.

[0056] The objectives, features, and advantages of this invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0057] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0058] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0059] In this invention, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined with "first" or "second" may explicitly or implicitly include at least one of that feature. Furthermore, the technical solutions of the various embodiments can be combined with each other, but only on the basis of being achievable by those skilled in the art. If the combination of technical solutions is contradictory or impossible to implement, such a combination of technical solutions should be considered non-existent and not within the scope of protection claimed by this invention.

[0060] Please refer to Figures 1 to 9 The present invention provides a real-time guidance method for concrete vibration operations, comprising the following steps:

[0061] S1, acquire a construction site image containing multiple pre-deployed fixed spatial reference points, and the world coordinates and pixel coordinates of each fixed spatial reference point; calculate spatial transformation parameters based on the pixel coordinates and the world coordinates; in a preferred example, site landmarks can be pre-selected, such as... Figure 4 The area outlined in purple requires operators to select at least four fixed spatial reference points based on the construction site image. These fixed spatial reference points must be immovable and easily identifiable, and should be distributed around the perimeter of the storage area. Furthermore, world coordinates can be obtained through measurement, while pixel coordinates can be directly read from the image.

[0062] S2, based on the spatial transformation parameters, the vibratory compaction area in the construction site image is mapped to generate a digital map with a true physical scale; the effective range of the vibrator is obtained, and a monitoring grid is divided on the digital map according to the effective range, such as... Figure 5 As shown;

[0063] S3, acquire the on-site video stream of the construction process, perform frame extraction processing on the on-site video stream according to a preset period and identify the current concrete surface, and then determine whether the current concrete surface meets the preset change conditions; if so, recalculate the spatial transformation parameters based on the fixed spatial reference point and update the digital map and monitoring grid.

[0064] S4, determine whether the vibrator in the image is in an effective vibration operation state; if so, determine the equivalent vibration point in the image, and convert the pixel coordinates of the equivalent vibration point to the world coordinates on the digital map according to the spatial transformation parameters (the spatial transformation parameters recalculated after step S3).

[0065] S5, map the equivalent vibration points after conversion (i.e., after processing in step S4) to the updated monitoring grid, and count the vibration coverage of each monitoring grid; compare the vibration coverage with the preset construction standard to identify under-vibrated or over-vibrated grid units; wherein, generate directional guidance information pointing to the under-vibrated grid unit for the under-vibrated grid unit; generate early warning information for the over-vibrated grid unit.

[0066] Specifically, if the construction workers encounter an over-vibrated grid cell during the vibration process, an instruction ("The current location has been over-vibrated, please move to the next area for vibration") can be sent to the construction workers. Furthermore, the system can count the under-vibrated or over-vibrated grid cells within the entire area and promptly push out guidance messages ("There is a missed / under-vibrated area xx, please move x meters in the x direction for vibration") to guide the workers to move to the corresponding grid for vibration.

[0067] Furthermore, by repeatedly executing the above steps until the concrete vibration construction of this layer is completed, a set of all equivalent vibration points is obtained. Based on the vibration time and grid division of each equivalent vibration point, a color cloud map is generated and displayed in a circular diffusion pattern. A heat map is then drawn and overlaid on the original base map, and simultaneously stored and displayed on the digital platform. Figure 9 As shown.

[0068] Specifically, each equivalent vibration point in the set of equivalent vibration points can be used as an interpolation node, and the vibration time of each equivalent vibration point can be used as an interpolation attribute. A continuous vibration time distribution surface can be generated within the concrete placement surface using a spatial interpolation algorithm. Calculations can be performed within the concrete placement surface using each equivalent vibration point in the set of equivalent vibration points as an interpolation node, its vibration time as an interpolation attribute, and a spatial interpolation algorithm (such as the inverse distance weighting method, Kriging method, or radial basis function method).

[0069] Based on the vibration time distribution surface, a color gradient is used for rendering to generate a vibration quality heatmap; as shown... Figure 9 The heatmap shown uses a gradient from blue to red, with redder areas representing longer vibration times and lighter areas (like blue) representing shorter times. This improves data interpretation efficiency and allows for instantaneous assessment of complex construction quality. Vibration time is a core quality control factor for ensuring concrete compaction; it cannot be too long or too short. Therefore, the heatmap's color-coding allows for a direct visual identification of under- or over-vibrated areas. Furthermore, the uniformity of color distribution on the heatmap helps workers assess the uniformity of concrete vibration.

[0070] The vibration quality heat map is overlaid and fused with a pre-stored concrete surface base map. The pre-stored concrete surface base map can be obtained by adjusting the zoom and magnification of a portable smart terminal (such as a monocular camera) before starting the algorithm, keeping it fixed during the acquisition process, and storing the first frame image of the concrete surface as the base map. The overlaid and fused result is then sent to the backend management system for storage and display.

[0071] This application's solution, based on a dynamic visual grid segmentation scheme using marker references, solves the industry-wide problem of visual reference failure caused by the rise of the concrete pouring surface. The grid automatically updates as the concrete surface grows, ensuring synchronization between the digital map, monitoring grid, and the concrete surface. This guarantees continuous accuracy in detection and guidance throughout the entire vibration cycle, providing a unique and reliable spatiotemporal reference for precise global control. Through instance segmentation models and conditional judgments, it can automatically and in real-time accurately identify whether the vibrator is in a valid operational state from complex video streams, reducing interference from invalid data. Furthermore, this invention innovatively employs a monocular vision combined with visual grid segmentation technology, replacing the expensive binocular vision system. Image acquisition can be performed using a single-lens camera, resulting in low cost and simple deployment, significantly lowering the barrier to technology promotion.

[0072] In a preferred embodiment, the step S2 of obtaining the effective range of the vibrator and dividing the monitoring grid on the digital map according to the effective range includes the following steps:

[0073] S21, obtain the effective radius of action of the vibrator, and determine the physical side length of the monitoring grid unit based on the effective radius of action; for example, the physical side length of the monitoring grid unit is twice the effective radius of action. In other embodiments, those skilled in the art can also make adaptive settings according to actual needs.

[0074] S22, calculate the scale parameter of the digital map, and calculate the pixel size corresponding to each monitoring grid according to the effective radius of action and the scale parameter; the scale parameter is the ratio of the distance on the map to the actual distance, which can be calculated using existing mature technologies, and will not be elaborated here.

[0075] S23, Based on the pixel size, grid lines are drawn on the digital map to form the monitoring grid. On the digital map image, starting from a preset origin, a straight line is drawn at every pixel size in both the horizontal and vertical directions, ultimately forming a crisscrossing grid line array covering the entire vibratory compaction chamber surface area, resulting in multiple monitoring grids, for example... Figure 5 As shown.

[0076] In a preferred embodiment, step S3 includes the following steps:

[0077] S31, acquire the on-site video stream of the construction process, extract frames from the on-site video stream according to a preset time interval, process the extracted images using a pre-trained first instance segmentation model to obtain the segmentation mask of the current concrete slab surface, and then extract the outline of the concrete slab surface from the segmentation mask; as an example, the preset time interval is set to 10 minutes.

[0078] As a preferred example, the first instance separation model is obtained through the following steps:

[0079] Collect images of on-site vibration compaction operations and construct a training dataset;

[0080] Each image in the training dataset is labeled using an annotation tool; specifically, the segmentation mask of the current concrete slab surface is labeled to generate corresponding instance segmentation mask labels, such as... Figure 3 The blue-purple area shown; the annotation tool can be the open-source annotation tool LabelMe. The generated instance segmentation mask labels annotate the precise contour information of the current concrete slab surface segmentation mask, so that the finally trained instance segmentation model can not only identify where there are labels, but also identify the corresponding shapes, thereby improving the final judgment accuracy.

[0081] The labeled training dataset is divided into training set, validation set and test set according to a preset ratio;

[0082] The instance segmentation neural network is trained using the training set, the parameters are tuned using the validation set, and the performance of the instance segmentation model is evaluated using the test set until the model converges or the performance on the validation set reaches a preset index, thus obtaining the first instance segmentation model after training.

[0083] S32, calculate the outline area based on the outline of the concrete sump surface, and calculate the intersection point between the outline of the concrete sump surface and the outline of the marker reference object associated with the fixed spatial reference point, and then obtain the pixel coordinates of the intersection point; the marker reference object is predetermined, the marker reference object will not be moved, and the marker reference object is distributed around the sump surface.

[0084] S33: Calculate the area change rate of the contour area compared to the previously updated contour area, and calculate the actual physical displacement of the pixel coordinates of the intersection points after transformation by the current spatial transformation parameters. Since the vibratory compaction surface rises during concrete pouring, it is adjusted in real time. In this embodiment, the first instance separation model is used for identification, followed by a comprehensive judgment in S34. The actual physical displacement can be the maximum value among multiple displacement values ​​or the maximum displacement in the height direction.

[0085] S34, determine whether the area change rate is greater than a first preset threshold, and determine whether the actual physical displacement of at least one of the intersection points is greater than a second preset threshold.

[0086] S35: When both of the following conditions are met simultaneously: the area change rate is greater than a first preset threshold, and the actual physical displacement at at least one of the intersection points is greater than a second preset threshold, the current concrete surface is determined to meet the preset change conditions. Only when both conditions are met simultaneously, i.e., when both area and position change significantly, is a substantial change in the surface determined, ensuring the necessity of each update. If at least one of the two conditions is not met, the current concrete surface is determined not to meet the preset change conditions, and the process returns to step S31 to obtain new data for further analysis and judgment.

[0087] S36. Based on the current concrete surface outline and the fixed spatial reference point, recalculate the spatial transformation parameters and update the digital map and monitoring grid based on the recalculated spatial transformation parameters.

[0088] This embodiment, based on dynamic visual grid segmentation technology using marker references, solves the industry problem of visual reference failure caused by the rise of the concrete pouring surface. The grid can automatically update as the pouring surface grows, ensuring continuous accuracy of detection and guidance throughout the entire vibration cycle. It provides a unique and reliable spatiotemporal reference for precise global control, ensuring that the updated digital map accurately reflects the concrete area of ​​the pouring surface.

[0089] In a preferred embodiment, step S4, determining whether the vibrator in the image is in an effective vibration operation state, includes the following steps:

[0090] The pre-trained second instance segmentation model is used to identify the segmentation mask of the vibrator operator and the segmentation mask of the vibrator rod in each frame of the image. The pre-trained second instance segmentation model is used to identify the input image and output the pixel-level segmentation mask. It accurately marks all pixels in the image that belong to the two independent instances of the vibrator operator segmentation mask and the vibrator rod segmentation mask, thereby obtaining the accurate shape of the target and providing a reliable data foundation for subsequent calculations.

[0091] The cross-union ratio (CUNR) of the vibration operator's segmentation mask and the vibration rod's segmentation mask in the identification results is calculated. When the CUNR is greater than a set threshold, the vibration rod detection is determined to be correct, and the vibration rod is determined to be held and operated normally by the worker. When the CUNR is less than or equal to the set threshold, the vibration rod detection result is filtered. The set threshold can be set based on experience according to the current layer surface construction, or determined through trial operation in the early stage of deployment of this application scheme. As a preferred example, the set threshold is 0.15.

[0092] Perform a geometric intersection operation between the vibratory rod segmentation mask and the segmentation mask of the current concrete placement surface. If there is an intersection between the vibratory rod segmentation mask and the segmentation mask of the current concrete placement surface, mark the current operation state as a valid vibration operation state; if there is no intersection between the vibratory rod segmentation mask and the segmentation mask of the current concrete placement surface obtained from the first instance segmentation model, mark the current operation state as an invalid vibration operation state.

[0093] As a preferred example, a closed surface can be constructed by combining the outer contour of the vibratory rod's dividing mask with the outer contour of the current concrete placement surface's dividing mask. The intersection area is then calculated. If the area of ​​the intersection area is greater than 0.3 times the area of ​​the vibratory rod's dividing mask, the current state is marked as a vibration operation state; otherwise, it is marked as an idle hovering state (i.e., an invalid vibration operation state). This step accurately determines whether the vibration point is in an effective operation state, avoiding the inclusion of invalid time in the operation duration. It solves the problem of relying on the visual observation and subjective experience of supervisors in existing technologies.

[0094] The training process of the second instance segmentation model is based on the same principle as the first instance segmentation model, and will not be repeated here. The segmentation mask for the vibrating personnel is as follows: Figure 2 The areas marked in red and green are where the vibrator divides the mask, as shown. Figure 2 The orange-yellow area in the middle.

[0095] Preferably, determining the equivalent vibration point in the image in step S4 includes the following steps:

[0096] When the current operation is in the effective vibration operation state, extract the contour points of the vibratory rod dividing the mask, and select the point at the bottom of the vibratory rod as the equivalent vibration point based on the contour points.

[0097] Specifically, the point at the bottom of the vibrator can be selected as the equivalent vibration point along the length of the mask divided by the vibrator, such as... Figure 8 The small green dots shown indicate the precise determination of the vibration point.

[0098] In a preferred embodiment, step S5, which maps the converted equivalent vibration points to the updated monitoring grid and calculates the vibration coverage of each monitoring grid, includes the following steps:

[0099] Using the converted equivalent vibration point as the center and the effective radius of the vibrator as the radius, a circular influence area is determined on the digital map, and the cumulative number of vibrations for each monitoring grid that falls into the circular influence area is increased by one.

[0100] It is worth noting that this application does not simply consider a single vibration operation as a point. Instead, it uses the equivalent vibration point as the center and the effective radius of the vibrator as the radius to determine a circular influence area on the digital map. This embodiment fully considers the principle of concrete vibration and the radius of influence of vibration. A single vibration operation affects a region, not a single point, thereby determining all monitoring grids affected by this vibration. A circular influence area may simultaneously cover multiple monitoring grids, thus achieving simultaneous counting of multiple monitoring grids in a single operation and accurately reflecting the actual coverage of the vibration effect.

[0101] Based on the final cumulative number of vibrations for each monitoring grid, and combined with a preset time conversion relationship, the cumulative vibration duration for each monitoring grid is calculated, and this cumulative vibration duration is used as the vibration coverage. By converting the cumulative number of vibrations into the cumulative vibration duration, and then accurately comparing it with the preset construction standards specified in the construction specifications, it is possible to accurately determine whether the vibration area is normal, or whether it is under-vibrated or over-vibrated.

[0102] Preferably, the preset time conversion relationship is: cumulative vibration time = cumulative vibration count × (K / F); where K is the number of original video frames between two adjacent frames extracted from the on-site video stream, and F is the number of frames per second of the on-site video stream. Through a mathematical conversion model based on the fundamental principle of frame extraction detection, the equivalent time of the continuous vibration operation process is accurately calculated and quantified, providing a reliable data foundation for quality assessment.

[0103] The present invention also provides a real-time guidance device for concrete vibration operation, used to perform the real-time guidance method for concrete vibration operation as described above, comprising:

[0104] The transformation parameter determination unit is used to acquire a construction site image containing multiple pre-laid fixed spatial reference points, as well as the world coordinates and pixel coordinates of each fixed spatial reference point; and to calculate spatial transformation parameters based on the pixel coordinates and the world coordinates.

[0105] The map grid unit is used to map the vibratory compaction area in the construction site image to generate a digital map with a real physical scale relationship according to the spatial transformation parameters; to obtain the effective range of the vibratory rod, and to divide the monitoring grid on the digital map according to the effective range;

[0106] The dynamic update unit is used to acquire on-site video streams during the construction process, perform frame extraction processing on the on-site video streams according to a preset period, identify the current concrete surface, and then determine whether the current concrete surface meets the preset change conditions; if so, the spatial transformation parameters are recalculated based on the fixed spatial reference point, and the digital map and monitoring grid are updated.

[0107] The vibration judgment unit is used to determine whether the vibrator in the image is in an effective vibration operation state; if so, it determines the equivalent vibration point in the image and transforms the pixel coordinates of the equivalent vibration point to the world coordinates on the digital map according to the spatial transformation parameters.

[0108] The guidance and decision-making unit is used to map the converted equivalent vibration points to the updated monitoring grid and to count the vibration coverage of each monitoring grid; based on the vibration coverage, it compares with the preset construction standard to identify under-vibrated or over-vibrated grid units; wherein, for under-vibrated grid units, it generates directional guidance information pointing to the under-vibrated grid units; and for over-vibrated grid units, it generates early warning information.

[0109] The present invention also provides a real-time guidance device for concrete vibration operation, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the real-time guidance method for concrete vibration operation as described above.

[0110] The present invention also provides a storage medium storing a computer program that, when executed by a processor, implements the steps of the real-time guidance method for concrete vibration operation as described above.

[0111] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0112] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A method for real-time guidance of concrete vibration operations, characterized in that, Includes the following steps: S1, acquire a construction site image containing multiple pre-deployed fixed spatial reference points, as well as the world coordinates and pixel coordinates of each fixed spatial reference point; calculate spatial transformation parameters based on the pixel coordinates and the world coordinates; S2, based on the spatial transformation parameters, the vibratory compaction area in the construction site image is mapped to generate a digital map with a real physical scale relationship; the effective range of the vibrator is obtained, and a monitoring grid is divided on the digital map according to the effective range; S3, acquire the on-site video stream of the construction process, perform frame extraction processing on the on-site video stream according to a preset period and identify the current concrete surface, and then determine whether the current concrete surface meets the preset change conditions; if so, recalculate the spatial transformation parameters based on the fixed spatial reference point and update the digital map and monitoring grid. S4, determine whether the vibrator in the image is in an effective vibration operation state; if so, determine the equivalent vibration point in the image, and convert the pixel coordinates of the equivalent vibration point to the world coordinates on the digital map according to the spatial transformation parameters. S5, map the converted equivalent vibration points to the updated monitoring grid, and count the vibration coverage of each monitoring grid; compare the vibration coverage with the preset construction standard to identify under-vibrated or over-vibrated grid units; wherein, generate directional guidance information pointing to the under-vibrated grid unit for the under-vibrated grid unit; generate early warning information for the over-vibrated grid unit. Step S3 includes the following steps: S31, acquire the on-site video stream of the construction process, extract frames from the on-site video stream according to a preset time interval, process the extracted images using a pre-trained first instance segmentation model to obtain the segmentation mask of the current concrete slab surface, and then extract the outline of the concrete slab surface from the segmentation mask. S32, calculate the outline area based on the concrete sump surface outline, calculate the intersection point between the concrete sump surface outline and the outline of the marker reference object associated with the fixed spatial reference point, and then obtain the pixel coordinates of the intersection point; S33, calculate the rate of change of the contour area compared to the previously updated contour area, and calculate the actual physical displacement of the pixel coordinates of the intersection point after being transformed by the current spatial transformation parameters. S34, determine whether the area change rate is greater than a first preset threshold, and determine whether the actual physical displacement of at least one of the intersection points is greater than a second preset threshold. S35, when both of the following conditions are met simultaneously: the area change rate is greater than the first preset threshold and the actual physical displacement with at least one of the intersection points is greater than the second preset threshold, the current concrete surface is determined to meet the preset change conditions. S36. Based on the current concrete surface outline and the fixed spatial reference point, recalculate the spatial transformation parameters and update the digital map and monitoring grid based on the recalculated spatial transformation parameters. The step S4, determining whether the vibrator in the image is in an effective vibration operation state, includes the following steps: The pre-trained second instance segmentation model is used to identify the segmentation mask of the vibrating personnel and the segmentation mask of the vibrating rod in each frame of the image. The cross-union ratio (CUNR) of the vibration personnel segmentation mask and the vibration rod segmentation mask in the identification results is calculated. When the CUNR is greater than a set threshold, the vibration rod detection is determined to be correct; when the CUNR is less than or equal to the set threshold, the vibration rod detection result is filtered. Perform a geometric intersection operation between the vibrator segmentation mask and the segmentation mask of the current concrete placement surface. If there is an intersection between the vibrator segmentation mask and the segmentation mask of the current concrete placement surface, mark the current operation state as a valid vibration operation state; if there is no intersection between the vibrator segmentation mask and the segmentation mask of the current concrete placement surface, mark the current operation state as an invalid vibration operation state.

2. The real-time guidance method for concrete vibration operation according to claim 1, characterized in that, Step S2, which involves obtaining the effective range of the vibrator and dividing the monitoring grid on the digital map based on the effective range, includes the following steps: S21, obtain the effective radius of action of the vibrator, and determine the physical side length of the monitoring grid unit based on the effective radius of action; S22, calculate the scale parameter of the digital map, and calculate the pixel size corresponding to each monitoring grid according to the effective radius of action and the scale parameter; S23, draw grid lines on the digital map to form the monitoring grid according to the pixel size.

3. The real-time guidance method for concrete vibration operation according to claim 1, characterized in that, Determining the equivalent vibration point in the image in step S4 includes the following steps: When the current operation is in the effective vibration operation state, extract the contour points of the vibratory rod dividing the mask, and select the point at the bottom of the vibratory rod as the equivalent vibration point based on the contour points.

4. The real-time guidance method for concrete vibration operation according to claim 2, characterized in that, Step S5, which maps the converted equivalent vibration points to the updated monitoring grid and calculates the vibration coverage of each monitoring grid, includes the following steps: Using the converted equivalent vibration point as the center and the effective radius of the vibrator as the radius, a circular influence area is determined on the digital map, and the cumulative number of vibrations for each monitoring grid that falls into the circular influence area is increased by one. Based on the final cumulative number of vibrations for each monitoring grid, the cumulative vibration duration for each monitoring grid is calculated using a preset time conversion relationship, and the cumulative vibration duration is used as the vibration coverage.

5. The real-time guidance method for concrete vibration operation according to claim 4, characterized in that, The preset time conversion relationship is: cumulative vibration duration = cumulative vibration times × (K / F); where K is the number of original video frames between two adjacent frames extracted from the on-site video stream, and F is the number of frames per second of the on-site video stream.

6. A real-time guidance device for concrete vibration operation, used to execute the real-time guidance method for concrete vibration operation as described in any one of claims 1-5, characterized in that, include: The transformation parameter determination unit is used to acquire a construction site image containing multiple pre-laid fixed spatial reference points, as well as the world coordinates and pixel coordinates of each fixed spatial reference point; and to calculate spatial transformation parameters based on the pixel coordinates and the world coordinates. Map grid cells are used to map the vibratory compaction area in the construction site image to generate a digital map with a true physical scale relationship based on the spatial transformation parameters. Obtain the effective range of the vibrator and divide the monitoring grid on the digital map according to the effective range; The dynamic update unit is used to acquire the on-site video stream of the construction process, perform frame extraction processing on the on-site video stream according to a preset period, identify the current concrete surface, and then determine whether the current concrete surface meets the preset change conditions. If so, the spatial transformation parameters are recalculated based on the fixed spatial reference point, and the digital map and monitoring grid are updated; The vibration judgment unit is used to determine whether the vibrator in the image is in an effective vibration operation state; if so, it determines the equivalent vibration point in the image and transforms the pixel coordinates of the equivalent vibration point to the world coordinates on the digital map according to the spatial transformation parameters. The guidance and decision-making unit is used to map the converted equivalent vibration points to the updated monitoring grid and to count the vibration coverage of each monitoring grid; based on the vibration coverage, it compares with the preset construction standard to identify under-vibrated or over-vibrated grid units; wherein, for under-vibrated grid units, it generates directional guidance information pointing to the under-vibrated grid units; and for over-vibrated grid units, it generates early warning information.

7. A real-time guidance device for concrete vibration operations, characterized in that, The method includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the real-time guidance method for concrete vibration operation as described in any one of claims 1 to 5.

8. A storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the real-time guidance method for concrete vibration operation as described in any one of claims 1 to 5.

Images