A concrete aggregate position tracking system and method based on X-ray binocular imaging

Abstract

The application discloses a concrete aggregate position tracking system and method based on X-ray binocular imaging, and the system comprises the following: a binocular X-ray imaging unit, an image processing and aggregate identification unit, an aggregate space positioning and track tracking unit, and a radiation safety protection unit; the application realizes visual representation of the concrete mixing process; the application realizes real-time, non-destructive and visual representation of the internal state of the concrete mixing process for the first time, breaks through the technical bottleneck of the concrete material "black box", and changes the concrete quality control from "post-inspection" to "real-time monitoring"; the process dynamic monitoring capability is comprehensively improved; the dynamic monitoring frequency of more than 10Hz is realized, millisecond-level mixing process changes can be captured, and aggregate position changes are recorded; the internal structure representation capability is broken through; the concrete interior can be penetrated, the aggregate distribution can be accurately positioned, and unevenness defects such as slurry aggregation zones and cavity zones can be identified.

Description

Technical Field

[0001] The present invention relates to the fields of concrete material science and engineering, process control and intelligent sensing technology, and in particular to a concrete aggregate position tracking system and method based on X-ray binocular imaging. Background Art

[0002] The uniformity of concrete aggregate directly determines the final performance of concrete. Its importance is reflected in three dimensions: (1) Mechanical performance: The three-dimensional spatial distribution of aggregate directly affects the load transfer path and stress distribution state of concrete. Uneven distribution will lead to structural defects such as stress concentration and weak interface transition zone, reducing compressive strength (decreasing by up to 30%) and flexural strength; (2) Durability: Differences in aggregate packing density will form permeation channels, accelerating chloride ion erosion (permeability coefficient increases by 2-3 orders of magnitude) and freeze-thaw damage (durability index decreases by more than 40%); (3) Construction performance: Uneven distribution of aggregate will significantly change the rheological properties of concrete, causing pumping blockage (increase in incidence by 5-8 times), pouring segregation and other engineering problems. Especially in modern high-strength concrete (water-cement ratio <0.3) and self-compacting concrete (extension >600mm), the precise control of aggregate particle size distribution (Dmax = 8-25mm) and volume fraction (60-75%) is more stringent. In traditional construction, relying on apparent test indicators such as slump cannot effectively evaluate the spatial distribution of aggregates, resulting in a 28-day compressive strength dispersion coefficient as high as 15-20%, which seriously restricts the controllability of concrete engineering quality.

[0003] However, due to the opacity of concrete, traditional testing methods cannot directly observe its internal state. As a result, the concrete mixing process has long been regarded as a "black box." Mixing quality relies primarily on empirical judgment and post-inspection, lacking scientific process monitoring and characterization methods. This opaque internal state of concrete severely restricts in-depth understanding of the mechanisms that form its working properties and also affects the real-time accuracy of quality control.

[0004] At present, the concrete working performance testing mainly includes the following methods:

[0005] 1. Traditional slump and flow tests: These methods, including slump, flow, expansion, L-box, and V-funnel tests, characterize the workability of a material by measuring flow time or geometric deformation. These methods are simple to use but only provide a single empirical parameter that cannot fully reflect the distribution and movement of aggregate within concrete. They also have large measurement errors (±10mm) and cannot identify internal defects.

[0006] 2. Shear-based rheological testing methods: Rotational rheometers such as the ICAR rheometer and Viskomat XL rheometer, as well as torque rheometers, can measure concrete's yield stress and plastic viscosity parameters. However, these devices can only test samples and are not representative of the entire batch of concrete, nor can they observe the actual movement of internal aggregates.

[0007] 3. AI-assisted non-contact characterization: Rheological properties are predicted by analyzing surface image videos during the mixing process. However, due to the opacity of concrete, these methods cannot obtain true information about the internal aggregate and can only infer surface properties.

[0008] These existing technologies all have obvious shortcomings: they are unable to monitor the distribution and position changes of aggregates inside concrete in real time, are unable to accurately obtain the movement trajectory of aggregates in three-dimensional space, and are unable to provide dynamic changes in aggregate distribution during concrete mixing.

[0009] In response to the above technical problems, the present invention provides a concrete aggregate position tracking system and method based on X-ray binocular imaging. Summary of the Invention

[0010] The present invention provides a concrete aggregate position tracking system and method based on X-ray binocular imaging, which solves the problems in the background technology.

[0011] In order to achieve the above object, the present invention adopts the following technical solutions:

[0012] A concrete aggregate position tracking system based on X-ray binocular imaging, the system comprising:

[0013] Binocular X-ray imaging unit, including two X-ray source-detector assemblies arranged at 90° perpendicular angles, used for real-time penetrating imaging of the concrete mixing process and obtaining X-ray projection images at different angles;

[0014] Image processing and aggregate recognition unit, used for X-ray image preprocessing, aggregate segmentation and recognition;

[0015] Aggregate spatial positioning and trajectory tracking unit, used to achieve three-dimensional spatial positioning and motion trajectory tracking of aggregate;

[0016] Radiation safety protection unit, used to ensure radiation safety of system operation.

[0017] Furthermore, the binocular X-ray imaging unit includes:

[0018] Two industrial-grade X-ray sources, operating at 300-450kV, for generating high-energy X-rays to penetrate concrete materials;

[0019] Two high-sensitivity flat-panel detectors with an acquisition frequency of ≥100 Hz and a spatial resolution of ≤0.5 mm, each in line with the corresponding X-ray source;

[0020] The concrete mixing container is made of special materials, such as carbon fiber composite materials, with an X-ray attenuation rate of less than 5%;

[0021] Image acquisition and transmission unit, including a high-speed image acquisition card and data storage device;

[0022] Radiation safety protection measures, including lead rooms, radiation dose monitoring devices and safety interlock mechanisms.

[0023] Furthermore, the image processing and aggregate identification unit includes:

[0024] X-ray image preprocessing module for flat-field correction, dynamic range adjustment, non-local means denoising, contrast-limited adaptive histogram equalization, and morphological reconstruction filtering;

[0025] The bone segmentation module is based on an improved U-Net deep learning network and is used to accurately segment bone areas from X-ray images;

[0026] The aggregate recognition and identification module is used to individually identify and identify the segmented aggregates and establish the corresponding relationship between the same aggregates in the horizontal and vertical projection images.

[0027] Furthermore, the aggregate spatial positioning and trajectory tracking unit includes:

[0028] Binocular imaging calibration module, used to perform geometric calibration of two X-ray imaging systems;

[0029] The 3D spatial positioning module realizes the 3D spatial positioning of aggregates based on polar line constraints and triangulation principles;

[0030] The trajectory tracking module realizes continuous tracking of aggregate motion trajectory based on multi-feature fusion and Kalman filter prediction;

[0031] Trajectory analysis module is used to analyze the movement characteristics and distribution status of aggregates.

[0032] Furthermore, the radiation safety protection unit includes:

[0033] Lead shielding room, the walls are made of lead plates with a thickness of not less than 5mm, ensuring that the outdoor radiation dose rate is less than 0.5μSv / h;

[0034] Multi-layer protection structure, including a local lead shield around the X-ray source and a lead backing plate on the non-working surface of the detector;

[0035] Interlock mechanism ensures that the X-ray source cannot be started when the shielding room door is open;

[0036] The radiation monitoring system monitors the ambient radiation level in real time and automatically alarms and cuts off the power supply of the X-ray source when the threshold is exceeded.

[0037] Another object of the present invention is to provide a concrete aggregate position tracking system method based on X-ray binocular imaging, comprising the following steps:

[0038] S1. Build an orthogonally arranged binocular X-ray imaging system to perform real-time penetrating imaging of the concrete mixing process from both horizontal and vertical directions.

[0039] S2. preprocessing the acquired X-ray projection image;

[0040] S3, using deep learning algorithms to perform aggregate segmentation and recognition on the preprocessed images;

[0041] S4, establishing a correspondence between the same aggregates in two orthogonal perspectives, and calculating the three-dimensional spatial coordinates of the aggregates by triangulation;

[0042] S5. Realize continuous tracking and analysis of aggregate movement trajectory;

[0043] S6. Aggregate distribution analysis;

[0044] S7. Radiation safety control.

[0045] Furthermore, the X-ray projection image preprocessing includes the following steps:

[0046] S11, flat field correction, to eliminate the effects of detector pixel response non-uniformity and X-ray light field non-uniformity;

[0047] S12, dynamic range adjustment, stretching image contrast through histogram analysis;

[0048] S13, non-local means denoising, suppressing quantum noise in X-ray images;

[0049] S14, contrast-limited adaptive histogram equalization to enhance the contrast of local areas;

[0050] S15, morphological reconstruction filtering, removes small-sized interference and irregular noise in the image.

[0051] Furthermore, the aggregate segmentation and identification in step S3 includes the following steps:

[0052] S31. Use the improved U-Net network structure to perform aggregate area segmentation;

[0053] S32, performing connected domain analysis on the segmented image to separate aggregates that are in contact with each other;

[0054] S33, extracting geometric features and grayscale features of each aggregate area;

[0055] S34. assigning a unique identifier to each identified aggregate;

[0056] S35. Based on the feature similarity, a correspondence relationship between the same aggregate in different projection views is established.

[0057] Furthermore, in step S4, the step of calculating the three-dimensional coordinates of the aggregate by triangulation includes:

[0058] S41. Perform geometric calibration on the two X-ray imaging systems using a standard calibration body;

[0059] S42. Based on the epipolar geometry theory, establish the correspondence between the aggregates in the two projection views;

[0060] S43. Calculate the three-dimensional coordinates of the aggregate using the triangulation principle;

[0061] S44. Use bundle adjustment technology to optimize the calculated three-dimensional coordinates to improve positioning accuracy.

[0062] Furthermore, in step S5, the steps of continuously tracking and analyzing the aggregate movement trajectory include:

[0063] S51. Construct aggregate feature descriptors by combining the spatial position, geometric characteristics and X-ray absorption properties of aggregates.

[0064] S52, using a Kalman filter algorithm to predict the position of the aggregate at the next moment;

[0065] S53, using the Hungarian algorithm to solve the data association problem of simultaneous tracking of multiple aggregates;

[0066] S54. Use a trajectory repair algorithm based on a motion model to handle the trajectory interruption problem caused by temporary occlusion of aggregates.

[0067] Furthermore, the aggregate distribution analysis in step S6 comprises the following steps:

[0068] S61. Statistical analysis of spatial distribution characteristics of aggregate at different times;

[0069] S62. Calculate the uniformity index of concrete and evaluate the mixing quality;

[0070] S63. Analyze the effects of aggregate shape, mixing speed and mixing time on aggregate distribution uniformity;

[0071] S64. Determine the optimal parameter combination for concrete mixing.

[0072] Furthermore, the radiation safety control in step S7 comprises the following steps:

[0073] S71. Check the closed state of the lead shielding room door before starting the X-ray source;

[0074] S72. Real-time monitoring of environmental radiation levels;

[0075] S73, automatically cut off the power supply to the X-ray source when an abnormal radiation level is detected or the safety interlock is triggered;

[0076] S74. Provide audible and visual warnings during system operation.

[0077] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0078] 1. Realize the visual representation of the concrete mixing process:

[0079] For the first time, it has achieved real-time, non-destructive, and visual representation of the internal state of the concrete mixing process, breaking through the technical bottleneck of the "black box" of concrete materials and transforming concrete quality control from "post-inspection" to "real-time monitoring."

[0080] The dynamic monitoring capability of the process is comprehensively improved: a dynamic monitoring frequency of more than 10Hz is achieved, which can capture millisecond-level changes in the mixing process and record changes in aggregate position.

[0081] A breakthrough has been achieved in the ability to characterize internal structures: the system can see through the interior of concrete, accurately locate aggregate distribution, and identify uneven defects such as slurry aggregation areas and void areas.

[0082] 2. The system design is highly innovative and practical:

[0083] The orthogonal arrangement design of binocular X-ray imaging has the advantages of simple structure, high imaging efficiency and high positioning accuracy. Compared with 360° circular scanning technology, it simplifies the system complexity, reduces costs and improves the system applicability.

[0084] The system integrates complete radiation safety protection measures to ensure safe operation and can be safely used in laboratories and industrial environments.

[0085] The system is compatible with a variety of concrete mix ratios, is applicable to a wide range of aggregate particle sizes (minimum identifiable particle size is 10mm), and has an aggregate volume fraction of 5%-40%, covering the vast majority of engineering application scenarios.

[0086] 3. Has extensive application expansion potential:

[0087] This technology is not only suitable for the research of concrete materials, but can also be extended to the research of various non-transparent materials such as soil, ceramic slurry, polymers, etc., and has broad application prospects.

[0088] The aggregate position data provided by the system can provide experimental verification for numerical simulation such as discrete elements, and improve the accuracy and reliability of the calculation model.

[0089] The technology provides a key sensing means for concrete digital twin technology, and provides technical support for the digital and intelligent transformation of concrete production.

[0090] Overall, the system solves the problem of the "black box" of the concrete mixing process, provides a revolutionary new tool for concrete material science research and engineering application, promotes the transformation of concrete technology from experience to science, improves the quality control level of concrete production and application, and produces significant economic and social benefits. BRIEF DESCRIPTION OF DRAWINGS

[0091] Figure 1 is the overall structure schematic diagram of the concrete aggregate tracking system based on X-ray binocular imaging of the present application;

[0092] Figure 2 is the arrangement schematic diagram of the binocular X-ray imaging system in the present application;

[0093] Figure 3 is the process schematic diagram of aggregate position recognition and trajectory extraction in the present application;

[0094] Figure 4 is the schematic diagram of the aggregate space positioning principle based on binocular X-ray in the present application. DETAILED DESCRIPTION

[0095] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0096] The overall architecture of the concrete aggregate position tracking system based on X-ray binocular imaging mainly includes the following parts:

[0097] 1. X-ray imaging system: including two sets of orthogonally placed X-ray source-detector groups, respectively from horizontal and vertical directions to penetrate and image the concrete mixing process.

[0098] 2. Image acquisition and transmission system: including a high-speed data acquisition card, an image cache system and a fiber transmission network, realizing high-speed acquisition and real-time transmission of X-ray images.

[0099] 3. Aggregate identification and tracking system: A deep learning-based image processing module that enables automatic identification, contour extraction, and position tracking of aggregates.

[0100] 4. Spatial positioning and data analysis system: Based on the principle of binocular imaging, the three-dimensional spatial positioning of aggregates is achieved through triangulation method, and statistical analysis of aggregate position data is performed.

[0101] 5. Radiation safety protection system: includes a multi-layer lead shielding structure, a radiation dose monitoring device and a safety interlock mechanism to ensure the radiation safety of the system operation.

[0102] The system works as follows: using two orthogonally placed X-ray source-detector sets to perform real-time penetrating imaging of the concrete mixing process, obtaining horizontal and vertical aggregate projection images respectively; identifying the outline and position of the aggregate through image processing algorithms; using binocular vision principles to perform triangulation and determine the three-dimensional spatial coordinates of the aggregate; and by continuously tracking the position changes of the aggregate, analyzing the motion trajectory and distribution characteristics of the aggregate, achieving visual analysis of the concrete mixing process.

[0103] X-ray imaging system

[0104] like Figure 2 As shown, the X-ray imaging system is the core module of the present invention, which mainly includes an X-ray source, an X-ray detector, a control unit and a mechanical bracket.

[0105] X-ray source

[0106] The X-ray source used in the present invention has the following characteristics:

[0107] 1. Energy range: 300-450kV, which can be adjusted according to the thickness and density of the concrete specimen to ensure that the X-rays can effectively penetrate the concrete sample.

[0108] 2. Focus size: ≤1.0mm to ensure the spatial resolution of imaging.

[0109] 3. Power: 3-5kW, providing sufficient radiation intensity to ensure image quality under high-speed sampling conditions.

[0110] 4. Stability: Output intensity fluctuation ≤1%, ensuring long-term working stability and consistency of image quality.

[0111] 5. Control system: Equipped with advanced electronic control system, it can accurately adjust the X-ray energy, dose rate and exposure time, and has overheating and overcurrent protection functions.

[0112] X-ray detectors

[0113] The X-ray detector used in the present invention has the following characteristics:

[0114] 1. Type: Adopts amorphous silicon flat-panel digital detector with high sensitivity, high dynamic range and low noise.

[0115] 2. Pixel size: ≤200μm, ensuring that the spatial resolution of the image can distinguish small-sized aggregates.

[0116] 3. Effective detection area: 30cm×30cm, enough to cover the cross-sectional area of ​​a standard concrete mixing container.

[0117] 4. Collection frequency: ≥100Hz, capable of capturing the high-speed movement of aggregates during concrete mixing.

[0118] 5. Dynamic range: ≥14 bits, meeting the needs of distinguishing different density components in concrete.

[0119] 6. Data output: Image data is transmitted to the image processing system in real time via a high-speed data interface (Camera Link / GigE Vision).

[0120] Orthogonal layout design

[0121] like Figure 3 As shown, the X-ray imaging system of the present invention adopts an orthogonal arrangement design, and the specific arrangement is as follows:

[0122] 1. Horizontal imaging system: X-ray source A is located on one side of the concrete mixing container, and detector A is located on the opposite side of the container. The central axes of the two are parallel to the horizontal plane, which is used to obtain the horizontal cross-sectional projection image of the concrete.

[0123] 2. Vertical imaging system: The X-ray source B is located directly above the concrete mixing container, and the detector B is located directly below the container. The central axes of the two are parallel to the vertical direction, which is used to obtain the vertical cross-sectional projection image of the concrete.

[0124] 3. Source-detector distance: The ratio of the source-to-object distance to the object-to-detector distance is 1:1.5. This ratio effectively reduces the impact of scattered radiation on image quality while ensuring imaging magnification.

[0125] 4. Synchronous control: The two imaging systems achieve precise synchronous exposure through a synchronous control unit, with a time synchronization accuracy better than 0.1ms, ensuring that the images in the two directions correspond to the internal state of the concrete at the same moment.

[0126] The advantage of this orthogonal arrangement design is that only two sets of X-ray source-detector systems are needed to achieve three-dimensional spatial positioning of aggregates. Compared with traditional CT imaging technology, it greatly simplifies the system structure, reduces cost and complexity, and improves the spatial resolution and temporal resolution of the system.

[0127] Aggregate image processing and recognition technology

[0128] Aggregate image processing and recognition is the key link in converting X-ray projection images into aggregate position information, which mainly includes three steps: image preprocessing, aggregate segmentation and aggregate recognition.

[0129] X-ray image preprocessing

[0130] X-ray image preprocessing aims to improve image quality and enhance the contrast between aggregate and background, and mainly includes the following steps:

[0131] 1. Flat-field correction: Using bright-field and dark-field images collected without a sample, flat-field correction is performed on the original X-ray image to eliminate the effects of detector pixel response non-uniformity and X-ray light field non-uniformity.

[0132] 2. Dynamic range adjustment: Through histogram analysis, the dynamic range of the image is adjusted to stretch the image contrast and highlight the grayscale difference between aggregate and paste.

[0133] 3. Non-local mean denoising: Using the non-local mean filtering algorithm, it effectively suppresses quantum noise in X-ray images while retaining the detailed information of aggregate edges.

[0134] 4. Contrast-limited adaptive histogram equalization: To address the problem of insufficient local contrast in X-ray projection images, contrast-limited adaptive histogram equalization (CLAHE) technology is used to enhance the contrast of local areas and highlight the edge features of aggregates.

[0135] 5. Morphological reconstruction filtering: Combining opening and closing operations with morphological reconstruction, it removes small-sized interference objects and irregular noise in the image, further improving the accuracy of aggregate recognition.

[0136] Aggregate segmentation

[0137] Aggregate segmentation is the process of extracting aggregate regions from preprocessed X-ray images. The present invention adopts an aggregate segmentation method based on deep learning, which specifically includes:

[0138] 1. Aggregate segmentation network: An improved U-Net network structure is used in combination with the ResNet feature extraction backbone network to build an end-to-end aggregate segmentation model.

[0139] 2. Data enhancement: Through data enhancement techniques such as rotation, scaling, and translation, the training samples are expanded to improve the generalization ability of the model.

[0140] 3. Multi-scale feature fusion: Design a feature pyramid structure to integrate feature information at different scales and improve the recognition ability of aggregates of various sizes.

[0141] 4. Attention mechanism: Introducing spatial attention module and channel attention module to enhance the model's attention to the aggregate area and improve segmentation accuracy.

[0142] 5. Aggregate edge optimization: Conditional random field (CRF) post-processing is used to optimize the aggregate segmentation boundary and improve the accuracy of aggregate contour extraction.

[0143] Aggregate identification and labeling

[0144] Individual identification and labeling of the segmented aggregate areas are performed. The main steps include:

[0145] 1. Connected domain analysis: Perform connected domain analysis on the segmented binary image to separate the aggregates that are in contact with each other.

[0146] 2. Aggregate feature extraction: Calculate the geometric features (area, perimeter, major axis, minor axis, eccentricity, etc.) and grayscale features (average grayscale, grayscale distribution, etc.) of each aggregate area.

[0147] 3. Aggregate identification: Assign a unique identifier to each identified aggregate and establish an aggregate identification database for subsequent aggregate tracking.

[0148] Aggregate matching: Based on the similarity of aggregate features, the correspondence between the same aggregates in the horizontal and vertical projection images is established, providing a basis for subsequent 3D positioning.

[0149] Aggregate binocular X-ray spatial positioning and trajectory tracking.

[0150] Three-dimensional spatial positioning of aggregate.

[0151] Based on orthogonal binocular X-ray projection images, the three-dimensional spatial positioning of aggregates is achieved. The main steps include:

[0152] 1. Camera calibration: Use a standard calibration body to perform geometric calibration on the two X-ray imaging systems to obtain intrinsic parameters (focal length, principal point coordinates, distortion coefficient, etc.) and extrinsic parameters (relative position and attitude).

[0153] 2. Epipolar constraint: Based on epipolar geometry theory, the correspondence between the aggregates in the two projection views is established to resolve the ambiguity of aggregate matching.

[0154] 3. Triangulation: Based on the known camera parameters and the coordinates of the aggregate on two projection planes, the three-dimensional spatial coordinates of the aggregate are calculated by the triangulation principle.

[0155] 4. Position optimization: Use bundle adjustment technology to optimize the initially calculated three-dimensional coordinates of the aggregate to improve positioning accuracy.

[0156] Aggregate trajectory tracking

[0157] Through time series association, continuous tracking of aggregate movement trajectory is achieved, mainly including:

[0158] 1. Multi-feature fusion tracking: The spatial position, geometric features, and X-ray absorption characteristics of the aggregate are combined to construct an aggregate feature descriptor for matching and associating aggregates at different times.

[0159] 2. Kalman filter prediction: Based on the historical movement trajectory of the aggregate, the Kalman filter algorithm is used to predict the position of the aggregate at the next moment, narrowing the search range and improving tracking efficiency.

[0160] 3. Multi-target association: The Hungarian algorithm is used to solve the data association problem of simultaneous tracking of multiple aggregates and handle complex situations such as aggregate intersection and occlusion.

[0161] Trajectory repair: To address the trajectory interruption problem caused by temporary occlusion of aggregates, a trajectory repair algorithm based on motion model is developed to achieve smooth connection of trajectories.

[0162] Radiation safety protection system

[0163] Considering the potential harm of X-rays to the human body, the present invention has designed a complete radiation safety protection system, which mainly includes:

[0164] 1. Lead shielding room: The entire X-ray imaging system is placed in a lead shielding room. The walls are made of lead plates with a thickness of not less than 5mm, ensuring that the outdoor radiation dose rate is less than 0.5μSv / h, in line with national radiation safety standards.

[0165] 2. Multi-layer protection structure: A local lead shield is set around the X-ray source and a lead back plate is set on the non-working surface of the detector to reduce the leakage of scattered radiation.

[0166] 3. Interlocking mechanism: A hardware interlocking device is set between the shielding room door and the X-ray source power supply to ensure that the X-ray source cannot be started when the door is open, preventing people from accidentally entering the radiation area.

[0167] 4. Radiation monitoring system: Radiation dose rate monitors are installed inside and outside the shielded room to monitor the environmental radiation level in real time. When the radiation dose exceeds the set threshold, it will automatically alarm and cut off the power supply of the X-ray source.

[0168] Warning signs: Obvious radiation warning signs are set up in the radiation area and at the entrance of the shielded room, and equipped with sound and light alarm devices to remind staff to pay attention to radiation safety.

[0169] Example 1: Aggregate distribution monitoring during concrete mixing

[0170] This example demonstrates the application of this system in monitoring aggregate distribution changes during concrete mixing.

[0171] The experiment adopts standard concrete mix ratio: cement 450 kg / m 3 , sand 830 kg / m 3 , gravel 1050 kg / m 3 , water 180 kg / m 3 , and the dosage of admixture is 1% of the mass of cement. The mixing equipment is a laboratory vertical forced mixer with a volume of 60 L.

[0172] The system parameters are set as follows: X-ray source voltage 400 kV, current 5 mA, focal point size 1.0 mm; detector pixel size 200 μm, acquisition frequency 100 Hz; image processing system sampling interval 0.1 s.

[0173] During the mixing process, the system captures the distribution state of the internal aggregate of the concrete in real time, calculates the uniformity index of the concrete through statistical analysis of the aggregate position, and evaluates the mixing quality. The results show that the system can accurately reflect the entire evolution process of the concrete from the initial non-uniform state to the final uniform state, providing a scientific basis for determining the optimal mixing time.

[0174] Example Two: Study on the Influence of Different Aggregate Shapes on Concrete Mixing Performance

[0175] This example demonstrates the application of the system in studying the influence of aggregate shape on concrete mixing performance.

[0176] In the experiment, three different shapes of coarse aggregate are selected: cubic gravel (with obvious edges and corners), sub-round pebbles (with moderate smoothness), and spherical ceramic particles (highly smooth), all with a particle size of 10-20 mm. Keeping other mix ratio parameters unchanged, concrete specimens of these three aggregates are prepared, and the system is used to monitor the mixing process.

[0177] By analyzing the spatial distribution and motion trajectory of different shaped aggregates during the mixing process, the results show that cubic aggregates are more likely to gather in the edge area of the mixing drum and have poor uniformity of distribution; spherical aggregates are more uniformly distributed throughout the container, and the shortest mixing time is required to reach a uniform state. This finding provides a new perspective for concrete mix design, i.e., optimizing aggregate shape can improve the mixing performance of concrete.

[0178] Example Three: Optimization of Mixing Parameters

[0179] This example demonstrates the application of the system in optimizing concrete mixing parameters.

[0180] The same concrete mixture ratio is used in the experiment, and different stirring speeds (15 rpm, 25 rpm, 35 rpm) and different stirring times (30 s, 60 s, 90 s, 120 s) are used to stir the concrete, and the distribution state of the aggregate in the concrete under each working condition is monitored by using the system.

[0181] By analyzing the uniformity index of aggregate distribution under different stirring parameters, the research results show that the stirring speed and the stirring uniformity are positively correlated, but when the speed exceeds a certain threshold (30 rpm in the experiment), the improvement of the uniformity is not obvious by continuously increasing the speed; the stirring time and the uniformity are also positively correlated, but there is a critical time point (90 s in the experiment), and the uniformity index tends to be stable after the time point. Based on these findings, the optimal parameter combination of concrete stirring can be determined, which can improve the production efficiency and reduce the energy consumption while ensuring the stirring quality.

[0182] The above is only the preferred specific embodiment of the present application, but the protection scope of the present application is not limited to this, any person skilled in the art can make equivalent replacement or change according to the technical scheme and the inventive concept of the present application within the technical range disclosed by the present application, which should be covered in the protection scope of the present application.

Claims

1. A concrete aggregate position tracking system based on X-ray binocular imaging, characterized by: The system includes the following: Binocular X-ray imaging unit, including two X-ray source-detector assemblies arranged at 90° perpendicular angles, used for real-time penetrating imaging of the concrete mixing process and obtaining X-ray projection images at different angles; Image processing and aggregate recognition unit, used for X-ray image preprocessing, aggregate segmentation and recognition; Aggregate spatial positioning and trajectory tracking unit, used to achieve three-dimensional spatial positioning and motion trajectory tracking of aggregate; Radiation safety protection unit, used to ensure radiation safety during system operation; The binocular X-ray imaging unit comprises: Two industrial-grade X-ray sources, operating at 300-450kV, for generating high-energy X-rays to penetrate concrete materials; Two high-sensitivity flat-panel detectors with an acquisition frequency of ≥100 Hz and a spatial resolution of ≤0.5 mm, each in line with the corresponding X-ray source; The concrete mixing container is made of special materials, such as carbon fiber composite materials, with an X-ray attenuation rate of less than 5%; Image acquisition and transmission unit, including a high-speed image acquisition card and data storage device; Radiation safety measures, including lead rooms, radiation dose monitoring devices, and safety interlock mechanisms; The image processing and aggregate identification unit includes: X-ray image preprocessing module for flat-field correction, dynamic range adjustment, non-local means denoising, contrast-limited adaptive histogram equalization, and morphological reconstruction filtering; The bone segmentation module is based on an improved U-Net deep learning network and is used to accurately segment bone areas from X-ray images; Aggregate recognition and identification module, used to identify and identify the segmented aggregates individually and establish the corresponding relationship between the same aggregates in the horizontal and vertical projection images; The aggregate spatial positioning and trajectory tracking unit includes: Binocular imaging calibration module, used to perform geometric calibration of two X-ray imaging systems; The 3D spatial positioning module realizes the 3D spatial positioning of aggregates based on polar line constraints and triangulation principles; The trajectory tracking module realizes continuous tracking of aggregate motion trajectory based on multi-feature fusion and Kalman filter prediction; Trajectory analysis module, used to analyze the motion characteristics and distribution status of aggregates; The radiation safety protection unit includes: Lead shielded room, with walls made of lead sheets; Multi-layer protection structure, including a local lead shield around the X-ray source and a lead backing plate on the non-working surface of the detector; Interlock mechanism ensures that the X-ray source cannot be started when the shielding room door is open; The radiation monitoring system monitors the ambient radiation level in real time and automatically alarms and cuts off the power supply of the X-ray source when the threshold is exceeded.

2. A method for tracking the position of concrete aggregates based on X-ray binocular imaging, characterized by: The method is implemented based on the concrete aggregate position tracking system based on X-ray binocular imaging according to claim 1, and the method comprises the following steps: S1. Build an orthogonally arranged binocular X-ray imaging system to perform real-time penetrating imaging of the concrete mixing process from both horizontal and vertical directions. S2. preprocessing the acquired X-ray projection image; S3, using deep learning algorithms to perform aggregate segmentation and recognition on the preprocessed images; S4, establishing a correspondence between the same aggregates in two orthogonal perspectives, and calculating the three-dimensional spatial coordinates of the aggregates by triangulation; S5. Realize continuous tracking and analysis of aggregate movement trajectory; S6. Aggregate distribution analysis; S7. Radiation safety control.

3. The method for tracking the position of concrete aggregates based on X-ray binocular imaging according to claim 2, wherein the X-ray projection image preprocessing comprises the following steps: S11, flat field correction, to eliminate the effects of detector pixel response non-uniformity and X-ray light field non-uniformity; S12, dynamic range adjustment, stretching image contrast through histogram analysis; S13, non-local means denoising, suppressing quantum noise in X-ray images; S14, contrast-limited adaptive histogram equalization to enhance the contrast of local areas; S15, morphological reconstruction filtering, removes small-sized interference and irregular noise in the image.

4. The method for tracking the position of concrete aggregates based on X-ray binocular imaging according to claim 2, characterized in that: Aggregate segmentation and identification in step S3 includes the following steps: S31. Use the improved U-Net network structure to perform aggregate area segmentation; S32, performing connected domain analysis on the segmented image to separate aggregates that are in contact with each other; S33, extracting geometric features and grayscale features of each aggregate area; S34. assigning a unique identifier to each identified aggregate; S35. Based on the feature similarity, a correspondence relationship between the same aggregate in different projection views is established.

5. The method for tracking the position of concrete aggregates based on X-ray binocular imaging according to claim 2, characterized in that: In step S4, the step of calculating the three-dimensional coordinates of the aggregate by triangulation includes: S41. Perform geometric calibration on the two X-ray imaging systems using a standard calibration body; S42. Based on the epipolar geometry theory, establish the correspondence between the aggregates in the two projection views; S43. Calculate the three-dimensional coordinates of the aggregate using the triangulation principle; S44. Use bundle adjustment technology to optimize the calculated three-dimensional coordinates to improve positioning accuracy.

6. The method for tracking the position of concrete aggregates based on X-ray binocular imaging according to claim 2, characterized in that: In step S5, the steps of continuously tracking and analyzing the aggregate movement trajectory include: S51. Construct aggregate feature descriptors by combining the spatial position, geometric characteristics and X-ray absorption properties of aggregates. S52, using a Kalman filter algorithm to predict the position of the aggregate at the next moment; S53, using the Hungarian algorithm to solve the data association problem of simultaneous tracking of multiple aggregates; S54. Use a trajectory repair algorithm based on a motion model to handle the trajectory interruption problem caused by temporary occlusion of aggregates.

7. The method for tracking the position of concrete aggregates based on X-ray binocular imaging according to claim 2, characterized in that: The steps of the aggregate distribution analysis in step S6 are as follows: S61. Statistical analysis of spatial distribution characteristics of aggregate at different times; S62. Calculate the uniformity index of concrete and evaluate the mixing quality; S63. Analyze the effects of aggregate shape, mixing speed and mixing time on aggregate distribution uniformity; S64. Determine the optimal parameter combination for concrete mixing.

8. The method for tracking the position of concrete aggregates based on X-ray binocular imaging according to claim 2, characterized in that: The radiation safety control in step S7 comprises the following steps: S71. Check the closed state of the lead shielding room door before starting the X-ray source; S72. Real-time monitoring of environmental radiation levels; S73, automatically cut off the power supply to the X-ray source when an abnormal radiation level is detected or the safety interlock is triggered; S74. Provide audible and visual warnings during system operation.

Images