Beam yard management system and method based on two-dimensional code technology
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-04-02
- Publication Date
- 2026-08-11
Smart Images
Figure CN122549461A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of smart construction technology, and in particular to a beam yard management system and method based on QR code technology. Background Technology
[0002] As the core structural component in a smart beam yard, the digital management of precast beams throughout their entire lifecycle is crucial. From rebar tying, concrete pouring, and steam curing to the most critical intelligent tensioning and grouting, each process not only requires accurate identification of the beam segment (ID), but also real-time and accurate recording of the execution status, duration, and physical location of each process to ensure the traceability of project quality and support the real-time updates of the digital twin (BIM) system.
[0003] Currently, smart beam yards mainly rely on the following technologies: First, perception technologies based on physical tags (such as RFID) and spatial positioning (such as RTK and UWB technologies) are used to determine the process by tracking the spatial location and trajectory of the precast beams. Although this method achieves beam identification and location perception, its process judgment logic is only based on spatial occupancy (that is, by mapping the spatial location to a specific process to identify the current process of the beam). It cannot determine whether the beam has actually performed the corresponding process, nor can it monitor the logical integrity of the process. In complex environments like beam yards, RFID technology uses radio signals to identify targets, but signal shielding is severe in the dense steel mesh environment. Existing QR code scanning methods are discrete recording methods; when a new process begins on the beam, workers still need to scan the code to report the current process status. If the process status is not updated in time, the backend will still see old data, leading to discrepancies between records and actual conditions and delays in digital twin updates. Secondly, while computer vision recognition technology detects targets in images or videos and judges process execution based on changes in the relative positions of precast beams and equipment, the complexity of construction sites prevents this solution from automatically linking process statuses and effectively monitoring quality throughout the entire process.
[0004] Therefore, how to eliminate the gap between precast beam identification and dynamic construction process logic in a complex and dynamic beam yard environment, effectively identify the true state of the process, and achieve full-process, fully automated judgment of the process is an urgent problem to be solved in the industry. Summary of the Invention
[0005] The purpose of this application is to provide a beam yard management system and method based on QR code technology to solve or alleviate the problems existing in the prior art.
[0006] To achieve the above objectives, this application provides the following technical solution: This application provides a beam yard management system based on QR code technology, including a perception layer, a processing layer, and an execution layer: The perception layer includes a laser deep etching unit and a video acquisition unit. The laser deep etching unit etches a QR code onto the surface of the precast beam using a laser to identify the logical identity ID of the precast beam and uses physical shadow features generated by the etching depth. The video acquisition unit is arranged in the work areas corresponding to different processes in the beam yard to collect real-time video streams from the work areas. The processing layer includes an edge computing unit, which runs an authentication module and a spatiotemporal logic state machine. The authentication module is configured to: recognize the QR code based on the real-time video stream, parse it to obtain the logical identity ID, extract the physical shadow features, and verify the authenticity of the precast beam's identity by comparing the physical shadow features with preset physical identifier attributes. The spatiotemporal logic state machine is configured to: establish a corresponding process verification instance for the precast beam after successful authentication, and determine the state transition based on the interaction between the construction equipment and the precast beam in the video stream. The execution layer includes a digital ledger module; the digital ledger module is communicatively connected to the edge computing unit; it receives a write instruction from the edge computing unit and persistently records it only when the time-space logic state machine determines that the current process has reached a preset process end state; the write instruction carries a timestamp and process completion evidence data.
[0007] Preferably, the process by which the identity verification module verifies the authenticity of the precast beam includes: The QR code region in the video frame is located and a sub-region of the image is extracted using an object detection algorithm. The image sub-region is decoded by calling the QR code decoding module to obtain the logical identity ID; The shadow feature analysis module is invoked to perform light and shadow gradient analysis on the image sub-region and extract the shadow feature vector generated by the etching depth. If decoding is successful and the shadow feature vector matches the preset physical characteristics, then the identity verification is deemed successful; if decoding is successful but the shadow feature vector does not match the preset physical characteristics, then there is a risk of identifier replacement and an alarm is generated.
[0008] Preferably, the state transition determination of the spatiotemporal logic state machine includes spatial topology constraints and time window constraints; The spatial topology constraint is used to determine the relative positional relationship between the construction equipment and the precast beam. When the relative positional relationship meets the preset geometric alignment conditions, a state transition is triggered. The time window constraint sets a minimum duration threshold and a maximum duration threshold for each state transition path. If the duration of the current state is less than the minimum duration threshold or greater than the maximum duration threshold, the spatiotemporal logic state machine determines that the process is invalid and resets, and does not perform the state transition operation.
[0009] Preferably, the tensioning process verification example of the spatiotemporal logic state machine includes an initial state, a beam position locking state, an equipment alignment state, a process execution state, and a process final state connected sequentially. The initial state is configured such that the beam QR code is not recognized or the work area is empty; The beam position locking status is triggered by the identity verification module confirming that the precast beam has entered the work area; The equipment alignment status is triggered when the geometric center of the construction equipment coincides with the anchorage area of the beam and the duration is greater than a first preset time. The execution status of the process is triggered by the recognition that the construction equipment is in working condition and the duration meets the process standard. The final state of the process is triggered by the detection that the construction equipment has detached from the beam.
[0010] Preferably, the edge computing unit also runs an anti-occlusion re-identification algorithm; If the QR code is obscured and cannot be recognized after identity verification, the anti-occlusion re-identification algorithm is configured to: extract irregular textures on the surface of the beam as temporary fingerprints and track the displacement of feature points of the temporary fingerprints; If the displacement of the feature point does not exceed the preset displacement threshold, the spatiotemporal logic state machine maintains the current process verification instance of the beam and continues to execute the determination of subsequent states.
[0011] Preferably, after receiving the write instruction, the digital ledger module binds and stores the timestamp, process completion evidence data, and logical identity ID to form an immutable process quality traceability record, and synchronously updates the beam status in the digital twin system.
[0012] This embodiment provides a beam yard management method based on QR code technology, applicable to the beam yard management system based on QR code technology provided in any of the above embodiments, characterized by including the following steps: The steps for identification and data acquisition are as follows: The laser deep etching unit etches the QR code onto the surface of the precast beam to identify the logical identity ID of the precast beam, and uses the physical shadow features generated by the etching depth; Video acquisition units are set up in the work areas corresponding to different processes in the beam yard, and the video acquisition units collect real-time video streams of the work areas; The data processing steps are as follows: Based on the real-time video stream, the QR code is identified, the logical identity ID is parsed, and the physical shadow features are extracted. The authenticity of the precast beam's identity is verified by comparing the physical shadow features with the preset physical identifier attributes. After the identity verification is passed, a corresponding process verification instance is established for the precast beam, and the state transition is determined based on the interaction state between the construction equipment and the precast beam in the video stream. The data recording steps are as follows: When the time-space logic state machine determines that the current process has reached the preset process end state, it sends a write instruction to the digital ledger module. The write instruction carries a timestamp and process completion evidence data. After receiving the write instruction, the digital ledger module binds and stores the timestamp, process completion evidence data and logical identity ID to form an immutable process quality traceability record, and synchronously updates the beam status in the digital twin system.
[0013] This embodiment provides a computer device, including a memory, a processor, and a computer program stored in the memory. The processor executes the computer program to implement the steps of the beam yard management method based on QR code technology described in the above embodiment.
[0014] This embodiment provides a computer-readable storage medium storing a computer program / instruction thereon, which, when executed by a processor, implements the steps of the beam yard management method based on QR code technology described in the above embodiment.
[0015] This embodiment also provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the beam yard management method based on QR code technology described in the above embodiment.
[0016] The technical solution of this application has the following beneficial effects: A laser-etched QR code system, combined with an identity verification module, utilizes the physical shadow characteristics generated by the etching depth for identification and verification, establishing a unique binding between the digital identity and the physical beam. A spatiotemporal logic state machine determines state transitions based on the interaction between construction equipment and the precast beam in the video stream, verifying the actual execution of processes. During writing, a write command is only sent to the digital ledger module when the spatiotemporal logic state machine determines that the current process has reached a preset final state, ensuring that only complete, qualified, and logically validated process data is permanently recorded, achieving consistency between the ledger and the actual data. In summary, through the perception layer, processing layer, and execution layer, a QR code-based digital management system for beam yards is constructed, effectively solving industry challenges such as difficulty in verifying process status and discrepancies between the ledger and the actual data, thereby improving the quality control capabilities and digitalization level of precast beam production. Attached Figure Description
[0017] Figure 1This is a flowchart illustrating a beam yard management method based on QR code technology according to some embodiments of this application.
[0018] Figure 2 This is a schematic diagram of a computer device. Detailed Implementation
[0019] The embodiments of this application will now be described with reference to the accompanying drawings.
[0020] Example 1 This embodiment provides a beam yard management system based on QR code technology, including a perception layer, a processing layer, and an execution layer.
[0021] The perception layer includes a laser deep etching unit and a video acquisition unit. The laser deep etching unit etches a QR code onto the surface of the precast beam to identify the logical identity ID of the precast beam and uses the physical shadow features generated by the etching depth. The video acquisition unit is arranged in the work areas corresponding to different processes in the beam yard to collect real-time video streams of the work areas. The processing layer includes an edge computing unit, which runs an authentication module and a spatiotemporal logic state machine. The authentication module is configured to: recognize QR codes based on real-time video streams, parse to obtain a logical identity ID, extract physical shadow features, and verify the authenticity of the precast beam's identity by comparing the physical shadow features with preset physical identifier attributes. The spatiotemporal logic state machine is configured to: establish corresponding process verification instances for the precast beam after successful authentication, and determine state transitions based on the interaction between the construction equipment and the precast beam in the video stream. The execution layer includes a digital ledger module; the digital ledger module is communicatively connected to the edge computing unit; the edge computing unit is configured to send a write instruction to the digital ledger module only when the time-space logic state machine determines that the current process has reached the preset process end state; the write instruction carries a timestamp and process completion evidence data.
[0022] In this embodiment, the beam yard management system adopts a layered architecture design, in which the perception layer, as the data acquisition front end of the system, is mainly responsible for the digital conversion of information from the physical world. The perception layer specifically includes a laser deep etching unit (e.g., a high-power fiber laser) and a video acquisition unit (e.g., a high frame rate industrial camera), which work together to achieve unique identification of the precast beam and continuous perception of the operation process.
[0023] The laser deep etching of QR codes involves using a high-power (preferably 50W~100W) fiber laser to emit a laser beam and perform deep etching on the surface of the precast beam (preferably the concrete protective layer or embedded steel plate on a non-load-bearing surface). The etching depth is preferably 0.5mm~2.0mm, forming a 3D code structure with a distinct physical texture.
[0024] The code element (module) of the QR code consists of an etched groove and an unetched surface. The groove can be filled with weather-resistant reflective material or retain the original concrete color to create a contrast.
[0025] It should be noted that the optimal timing for setting the laser deep etching QR code is after the precast beam has been demolded and the concrete strength has reached a preset threshold (e.g., 80% of the design strength), but before the intelligent tensioning process begins. At this point, the beam structure is stable and capable of bearing the laser processing, and it ensures that identity binding is completed before subsequent critical process monitoring.
[0026] In this embodiment, laser deep etching technology can laser-etch deep QR codes at specific locations on the surface of precast beams (or on pre-embedded metal nameplates). The etching depth can reach 0.5mm~2.0mm, and it is weather-resistant, corrosion-resistant, and resistant to alteration. Even after the beam body has been exposed to sunlight or subjected to transportation friction for a long time, the markings remain clearly readable. This not only alleviates the problem of RFID being blocked in the complex environment of the beam yard, but also avoids to some extent the problem of traditional QR codes being unreadable without damage.
[0027] In this embodiment, the laser deep-etched QR code provides a logical identity ID for the precast beam and physical shadow features based on the etching depth, giving it both data encoding and physical anti-counterfeiting attributes. In other words, the laser deep-etched QR code is not only a carrier of data information but also an anti-counterfeiting identifier for the physical entity. Through the physical structure and optical properties of the etching, it simultaneously provides the system with a logical identity ID and physical shadow features, thereby achieving a strong binding between digital identity and physical entity.
[0028] The logical identity ID serves as a unique identifier for the beam structure. When an industrial camera captures a QR code image, the scanning and decoding module identifies the distribution pattern of the code elements to reconstruct the binary data stream, thereby obtaining the logical identity ID. This is the function of the QR code as a data carrier, used to achieve digital indexing of the beam structure.
[0029] Unlike traditional QR codes, in this embodiment, the laser-etched QR code has a physically textured 3D code structure to provide physical shadow features. Specifically, due to the depth of the grooves in the image, the rate of change (gradient) of grayscale values at the edge of the grooves is significantly higher than that of the ink edge of a planar printed QR code, exhibiting a certain grayscale gradient distribution. Simultaneously, the direction of the shadow has a fixed geometric correspondence with the position of the light source, and presents a specific parallax effect as the camera's viewing angle changes, forming edge sharpness and shadow projection direction. When the industrial camera acquires the image, these optical features are captured simultaneously, serving as the basis for verifying the physical authenticity of the QR code.
[0030] The laser deep etching QR code provides two types of features for the processing layer through the above mechanism: the first layer of verification is the logic layer. By obtaining the logical identity ID, it confirms that the precast beam with the ID exists and is recorded in the background. Through the physical shadow feature, it confirms that the logical identity ID is bound to the real and unique precast beam, so as to prevent the logical identity ID from being misappropriated to other beams for data entry, which would cause the process data to be inconsistent with the actual situation.
[0031] In this embodiment, the video acquisition unit is deployed in the work areas corresponding to different processes in the beam yard to acquire real-time video streams from these work areas. Specifically, the video acquisition unit uses a high frame rate industrial camera, which serves as the core data acquisition terminal of the perception layer, responsible for converting the physical processes in the beam yard work area into digital video streams that can be processed by the edge computing unit. The high frame rate industrial camera can be a global shutter CMOS industrial camera with a frame rate greater than 60 fps, capable of capturing rapid actions such as the start-up of the tensioning jack oil pump and the insertion and removal of grouting pipes, avoiding the loss of key frames due to motion blur.
[0032] Furthermore, the high frame rate industrial camera adopts the deployment principle of process node directional coverage. The deployment nodes can include: the support directly above or to the side of the tensioning platform, the side above the grouting operation area, key nodes of the beam yard passage, etc., to ensure that the captured video can reflect the overall position of the beam, the transfer trajectory, and the area occupancy status.
[0033] After the high frame rate industrial camera is installed, the camera is calibrated to establish a mapping relationship between the image pixel coordinate system and the world coordinate system. Then, computer software is used to draw a virtual polygon area in the video screen, which is defined as the electronic fence of the work area. The position of the camera is recorded, and the unique identifier of the camera is associated with the process type of the work area.
[0034] In this embodiment, the processing layer serves as the on-site decision-making center of the beam yard management system. It is responsible for real-time intelligent analysis of the raw data collected by the perception layer and for outputting reliable instructions on the process status to the execution layer. The processing layer is deployed in the edge computing unit on the beam yard site, reducing network latency through localized computing to ensure the real-time performance and security of process status determination.
[0035] For example, the edge computing unit can be configured as an embedded industrial-grade computing device equipped with an artificial intelligence accelerator card (such as a GPU / NPU) and deployed in a protective cabinet at the beam yard site.
[0036] In this embodiment, an authentication module and a spatiotemporal logic state machine run within the edge computing unit. The authentication module, running on the AI accelerator card of the edge computing unit, ensures the authenticity of the precast beams entering the system, preventing identifier replacement or forgery. The spatiotemporal logic state machine is the process logic engine of the processing layer, used to model and verify the process execution of the precast beams.
[0037] Furthermore, the identity verification module is configured to: recognize the QR code based on the real-time video stream, parse it to obtain the logical identity ID, and extract physical shadow features. By comparing the physical shadow features with preset physical identifier attributes, the authenticity of the precast beam's identity is verified. In other words, the input to the identity verification module is a real-time video frame from a high-frame-rate industrial camera. It extracts the QR code from the video frame and parses the logical identity ID contained within it. Simultaneously, the identity verification module also extracts the physical shadow features from the video frame, compares them, and outputs a signal indicating successful or failed verification.
[0038] In one embodiment, the process by which the authentication module verifies the authenticity of the precast beam includes: The QR code region in the video frame is located using an object detection algorithm, and a sub-region of the image is extracted. The QR code decoding module is called to decode the sub-region of the image to obtain the logical identity ID. The shadow feature analysis module is called to perform light and shadow gradient analysis on the sub-region of the image and extract the shadow feature vector generated by the etching depth. If the decoding is successful and the shadow feature vector conforms to the preset physical characteristics, the identity verification is deemed successful. If the decoding is successful but the shadow feature vector does not conform to the preset physical characteristics, the risk of identity replacement is determined and an alarm is generated.
[0039] The target detection algorithm is a pre-trained target detection model (preferably YOLOv8 or Faster R-CNN architecture). This model is trained on QR code samples in a beam yard setting and can distinguish between laser-etched QR codes, ordinary printed QR codes, and other background information. During the precast beam production process, the target detection algorithm infers from the input video frames and outputs the bounding box coordinates of the detected QR code targets, such as coordinate information represented by a BBox. The authentication module further extracts a sub-region (ROI) containing the QR code from the original video frame based on the bounding box coordinates. To ensure the integrity of edge shadow features, the extraction range is extended outward by a preset number of pixels (preferably 10%~20%) from the bounding box to avoid the loss of etched shadows at the edges of the QR code. The extracted sub-region is then simultaneously sent to the QR code decoding module and the shadow feature analysis module.
[0040] The QR code decoding module calls a standard decoding library (such as ZBar, OpenCV decoding module, or ZXing) to perform grayscale, binarization, and error correction on the captured image sub-region. It then parses the encoded binary data stream from the QR code and converts it into the logical identity ID of the precast beam (e.g., the string "BEAM-2023-001"). If decoding fails (e.g., due to severe damage or incorrect encoding), the module directly outputs a decoding failure signal, terminating the authentication process for the current frame.
[0041] The shadow feature analysis module performs light and shadow gradient analysis on image sub-regions. Due to the depth difference (0.5mm~2.0mm) between the etched grooves and the plane, a specific transition between light and dark will be formed under illumination. This module quantifies this feature by calculating the gray-level gradient distribution of the image sub-regions. Specifically, the Sobel operator or Canny edge detection algorithm is used to calculate the gradient magnitude and gradient direction of each pixel in the image sub-region. Then, the QR code area is divided into a regular-sized grid (such as a 3×3 grid), and the gradient direction histogram and average gradient magnitude within each grid are statistically analyzed. The statistical features of all grids are concatenated to form the shadow feature vector of the QR code. Since laser-etched QR codes have rich high-frequency gradient components and high edge sharpness, while flat-printed or photographic QR codes exhibit low-frequency smooth transitions, some physical features can be preset. For example, if all average gradient magnitudes are greater than a preset threshold and the direction entropy is greater than a preset entropy value, the shadow feature vector is determined to meet the preset physical features. Otherwise, it does not meet the preset physical features.
[0042] Finally, the shadow feature analysis module determines the identity based on the similarity score between the results output by the QR code decoding module and the output of the shadow feature analysis module: If the QR code decoding module fails to decode, the identity is determined to be unknown and an alarm is generated; if the QR code decoding module successfully decodes but the shadow feature vector does not conform to the preset physical characteristics, it is determined that there is a risk of identifier replacement and an alarm is generated.
[0043] If the QR code decoding module successfully decodes the code and the shadow feature vector matches the preset physical features, that is, it can decode the logical identity ID from the image sub-region and the similarity score meets the preset score threshold range, then the identity verification is successful and the digital ledger module automatically records the logical identity ID. Based on the camera position of the current video frame, it automatically records the current process type, timestamp, and other information of the beam without requiring workers to scan the code and report.
[0044] When the system determines that the identity is unknown or there is a risk of identity replacement, it executes the following alert procedure: Write the current video frame, feature vector data, and judgment result into the security log area of the edge computing unit and mark it as a suspicious event. Send real-time messages to the management backend via the communication interface, including camera location and risk type; State machine locking prevents the creation of spatiotemporal logic state machine instances for this beam, thus preventing false process data from being written into the digital ledger.
[0045] Through the above steps, even if relevant personnel obtain a legitimate precast beam ID and generate a QR code, if the deep shadow features of laser etching cannot be replicated, the system can still identify the anomaly and issue an alarm, ensuring the authenticity of the beam yard management data.
[0046] After identity verification is successful, in this embodiment, the spatiotemporal logic state machine establishes a corresponding process verification instance for the precast beam and performs state transition determination based on the interaction state between the construction equipment and the precast beam in the video stream.
[0047] Specifically, the spatiotemporal logic state machine adopts an object-oriented design approach, assigning an independent process verification instance to each authenticated precast beam in each specific process.
[0048] In other words, the creation of process verification instances depends on the output of the authentication module. Only when the authentication module completes the dual verification of the logical identity ID and physical shadow features of the beam QR code in a certain video frame and outputs a signal that the authentication is successful will the edge computing unit trigger the instance creation process of the spatiotemporal logic state machine to ensure that all objects entering the process monitoring logic are physically real beams and that the QR codes are not misappropriated or impersonated.
[0049] After creation is triggered, the spatiotemporal logic state machine allocates memory space for the process verification instance and initializes it, including assigning a unique instance number to the instance, binding the precast beam logical identity ID, setting the current process type (such as tensioning, duct grouting), and the current state.
[0050] First, define the action sequences of the corresponding finite state machines for different processes. In this embodiment, a state series consisting of "ID-action-time" triples is designed, where the action sequence is denoted as... , This refers to the number of states in this process.
[0051] For example, for the tensioning process, the process verification instance of the spatiotemporal logic state machine includes an initial state, a beam position locking state, an equipment alignment state, a process execution state, and a process final state connected sequentially. The initial state is configured as follows: no beam QR code is detected or the work area is empty; the beam position locking state is triggered by the identity verification module confirming that the precast beam has entered the work area; the equipment alignment state is triggered by recognizing that the geometric center of the construction equipment coincides with the anchorage area of the beam and the duration is greater than a first preset time; the process execution state is triggered by recognizing that the construction equipment is in a stressed working state and the duration meets the process standard; the process final state is triggered by recognizing that the construction equipment has detached from the beam and that a physical mark remains at the end of the beam.
[0052] For example, the tensioning sequence is denoted as: ,in: This is the initial state. The beam is in a locked position. This indicates the equipment is in alignment. This indicates the process execution status. This represents the final state of the process.
[0053] In this embodiment, a process isolation mechanism is used to manage the quality of different processes. Specifically, for the same precast beam, when it enters different process areas, the system creates different instances. For example, the beam with ID BEAM-2023-001 first enters the tensioning platform. The spatiotemporal logic state machine creates instance A (type: tensioning) for this beam, and the state machine runs the tensioning process logic (S0→S4). When instance A reaches the final state S4 and is written into the digital ledger, the instance's lifecycle ends (memory is released), indicating that the tensioning process has been successfully completed and meets the quality requirements. The beam then enters the grouting area, and the system creates instance B (type: grouting). The state machine runs the grouting process logic (S0'→Sn'). When instance B reaches the final state Sn' and is archived, the instance's lifecycle ends, indicating that the grouting process has been successfully completed and meets the quality requirements. Instance B and instance A are independent and do not interfere with each other, but they are both associated with the same beam (both with ID BEAM-2023-001). In large beam yards, edge computing units may simultaneously manage multiple process instances of multiple beams. The system employs a multi-task scheduling mechanism, allocating independent memory space and time slices to each active instance to ensure that the states of each beam do not interfere with each other.
[0054] After each process verification instance is created, the spatiotemporal logic state machine performs a state transition determination based on video interaction. Specifically, the state transition determination of the spatiotemporal logic state machine includes spatial topology constraints and time window constraints. Spatial topology constraints are used to determine the relative positional relationship between the construction equipment and the precast beam. When the relative positional relationship meets the preset geometric alignment conditions, a state transition is triggered. The time window constraints set a minimum duration threshold and a maximum duration threshold for each level of state transition path. If the duration of the current state is less than the minimum duration threshold or greater than the maximum duration threshold, the spatiotemporal logic state machine determines that the process is invalid and resets, without performing a state transition operation.
[0055] The edge computing unit analyzes the video stream in real time using object detection algorithms and extracts the following features based on the analysis results: spatial relationships, motion features, and temporal duration features. Spatial relationships include, for example, the intersection-over-unity (IoU) ratio or center-point distance between the bounding boxes of equipment (such as jacks) and the bounding boxes of beam anchors; motion features can be the motion vectors of the equipment, such as the equipment's trajectory; and temporal duration features refer to the duration of the spatial relationships or the duration of the motion trajectory.
[0056] To accurately execute state transitions, the spatiotemporal logic state machine also receives real-time data uploaded by field sensors, such as pressure values acquired by pressure sensors and temperature values uploaded by temperature sensors.
[0057] After acquiring the data, for the current video frame, the spatiotemporal logic state machine queries the transition condition table corresponding to the current state. If the spatial topology constraints and time window constraints are met, the corresponding state transition operation is executed. The spatial topology constraints can be determined by the distance between the center points of the construction equipment and the precast beam being less than a set distance threshold, or the intersection-over-union (IoU) ratio being greater than a preset IoU threshold, thus satisfying the preset geometric alignment conditions. The time window constraint sets a minimum time window for each transition path. and maximum time window If the time for a certain process is less than the minimum required by the process... If the process is deemed invalid, it will not be recorded, nor will a state transition operation be performed.
[0058] Taking the tensioning process as an example: S0→S1 transition: When the spatial relationship shows that the beam enters the electronic fence of a certain work area, the instance transitions from the initial state S0 to the beam position locked state S1.
[0059] S1→S2 transition: When the spatial relationship shows that the jack (construction equipment) coincides with the center of the beam anchor and the duration is greater than the first preset time (e.g., 10 seconds), the instance transitions from the beam position locking state S1 to the equipment alignment state S2.
[0060] S2→S3 transition: When the target recognition algorithm shows that the jack piston displacement is >0 and the sensor data shows that the pressure value is >th threshold and the duration meets the process standard, the instance is transferred to the process execution state.
[0061] S3→S4 transition: When the spatial relationship shows that the jack has detached from the beam, the instance transitions to the final state of the process.
[0062] In this embodiment, dual constraints of spatial topology and time windows ensure that only genuine, complete, and process-standard operations can trigger state transitions, while simultaneously recording the time of state transitions. Each process has an independent state machine instance, avoiding state confusion between different processes (e.g., preventing misjudging tensioning state as grouting state) and ensuring the clarity of process logic. The entire lifecycle data of each instance, from creation to archiving, is recorded, forming a complete chain of process evidence. When full-process quality supervision is required, by querying the logical identity ID of a precast beam, all process instances and their state transition history experienced by the beam can be traced back.
[0063] Furthermore, to prevent temporary occlusion of the QR code during the precast beam process from causing process recording failure and to ensure the continuity of process monitoring, the edge computing unit also runs an anti-occlusion re-identification algorithm. When the spatiotemporal logic state machine enters the beam position locking state, if the QR code is occluded and cannot be identified, the anti-occlusion re-identification algorithm is configured to: extract the irregular texture on the beam surface as a temporary fingerprint and track the displacement of the feature points of the temporary fingerprint; if the displacement of the feature points does not exceed the preset displacement threshold, the spatiotemporal logic state machine maintains the current process verification instance of the beam and continues to execute the subsequent state determination.
[0064] If the authentication module fails to decode the QR code information for N consecutive frames (e.g., N=5 frames, approximately 0.1 seconds), it is determined that the QR code is obstructed, and the anti-obstruction re-identification process is immediately triggered. When the QR code is re-exposed and successfully decoded, and the decoded ID matches the ID of the temporary fingerprint tracking, the algorithm exits, and the system resumes authentication mode.
[0065] In this embodiment, after the QR code is obscured, a temporary fingerprint is constructed based on the natural irregular texture of the precast beam's concrete surface. The concrete surface has randomly distributed aggregates, pores, and color differences at the microscopic level. These features are unique and stable, and can serve as a physical characteristic identifier for the beam.
[0066] Specifically, while locking the beam ID in state S1, M feature-rich sub-regions are selected as tracking ROIs in the area surrounding the QR code (excluding the QR code itself). The preferred regions are the surface of the beam's side web or flange, avoiding areas with oil stains or damage. Then, the edge computing unit calls feature extraction algorithms (such as ORB, SIFT, or AKAZE) to process these regions, extracting key points, calculating feature descriptors for each key point, and combining the feature descriptors of all key points into a set, forming a temporary fingerprint database for the beam.
[0067] During the period when the QR code is obscured, feature points are continuously tracked through video frame sequences and their positions are compared with feature points in a temporary fingerprint database to determine whether the beam has undergone an identity change. If the displacement of the feature point does not exceed a preset displacement threshold, it indicates that the beam has not undergone an identity change, and the spatiotemporal logic state machine continues to record the corresponding state transition process using the verified identity ID in the state sequence. Otherwise, the current process verification instance is exited, the beam process abnormality is recorded in the log, and an alarm is triggered.
[0068] In this embodiment, the execution layer includes a digital ledger module; the digital ledger module is communicatively connected to the edge computing unit and is used to record production and management process data of the precast beams.
[0069] Specifically, the digital ledger module is responsible for receiving process judgment results from the processing layer and permanently storing them in the digital ledger module. Through a strict write trigger mechanism and evidence chain retention, the authenticity, immutability, and traceability of beam yard production data are ensured, completely resolving the discrepancy between the ledger and the actual situation caused by delayed manual reporting or lack of logical verification.
[0070] The digital ledger module is deployed on the server or in a private cloud environment at the beam yard management center, and communicates with the edge computing units deployed on site via wired network (fiber optic) or wireless private network (5G / WiFi6). The beam yard management center also deploys a digital twin system, which can read the data in the digital ledger module and display the corresponding status in real time in the twin.
[0071] The digital ledger module uses an append-only mode to write data. Once each process record is written, a unique hash value is generated and linked with the hash value of the previous record to form a chain structure. Modifications to historical data will cause the hash chain to break, which will be detected by the system, thus ensuring the immutability of process records.
[0072] Furthermore, a write command is sent to the digital ledger module only when the spatiotemporal logic state machine determines that the current process has reached a preset process completion state; the write command carries a timestamp and process completion evidence data. In other words, the edge computing unit is configured to restrict the timing of data writing, initiating a write request only when the spatiotemporal logic state machine determines that the current process has reached a preset process completion state S4.
[0073] When the process reaches its final state S4, the edge computing unit reads the data of the process verification instance from the local cache, encapsulates it into a data packet (e.g., using a JSON structure), generates a write instruction, and calls the write interface (API) of the digital ledger module. The mechanism of triggering write at the final state of the process ensures that only complete and qualified processes are recorded, avoiding the contamination of the ledger by semi-finished product data and achieving consistency between the ledger and the actual situation.
[0074] Furthermore, after receiving the write instruction, the digital ledger module binds and stores the timestamp, process completion evidence data, and logical identity ID to form an immutable process quality traceability record, and synchronously updates the beam status in the digital twin system.
[0075] For example, upon successful writing, the digital ledger module pushes the process completion event to the digital twin system via a message queue. The digital twin system then updates the virtual beam's status (e.g., "tensioning complete") to achieve real-time mapping between the physical beam yard and the virtual model. Simultaneously, process completion data is synchronized to the digital twin system in real time, enabling management to instantly grasp the beam yard's production progress and support refined scheduling and decision-making. In summary, the beam yard management system based on QR code technology provided in this embodiment employs laser deep etching of QR codes combined with an identity verification module. It utilizes the physical shadow characteristics generated by the etching depth for identity identification and verification. This technology is unaffected by the electromagnetic shielding of the dense steel mesh in the beam yard, resulting in higher signal stability. Through a dual verification mechanism combining logical identity ID and physical shadow characteristics, it can distinguish genuine laser-etched codes from QR codes printed on a flat surface, photographed, or displayed on a screen, preventing QR code replacement, misappropriation, or forgery. This ensures a unique binding relationship between digital identity and the physical beam structure, guaranteeing the authenticity of management data. The spatiotemporal logic state machine provided in this embodiment ensures the logical integrity of process execution, resolving the technical pain point of inconsistency between spatial occupancy and process execution, and achieving automated and intelligent judgment throughout the entire process flow. Furthermore, the write instructions received by the digital ledger module carry timestamps and process completion evidence data (such as keyframe videos), forming an immutable chain of evidence that meets the traceability requirements of the engineering quality responsibility system.
[0076] Example 2 This embodiment provides a beam yard management method based on QR code technology. This method is applied to the beam yard management system based on QR code technology described in any of the above embodiments, and includes: Step S1, the steps of identification and data acquisition: The laser deep etching unit etches the QR code onto the surface of the precast beam to identify the logical identity ID of the precast beam, and based on the physical shadow features generated by the etching depth; Video acquisition units are arranged in the work areas corresponding to different processes in the beam yard, and the video acquisition units collect the real-time video stream of the work area; Step S2, data processing steps: Based on the real-time video stream, identify the QR code, parse to obtain the logical identity ID, and extract the physical shadow features. By comparing the physical shadow features with the preset physical identifier attributes, verify the authenticity of the precast beam's identity. After the identity verification is passed, establish a corresponding process verification instance for the precast beam, and determine the state transition based on the interaction state between the construction equipment and the precast beam in the video stream. Step S3, the data recording steps: Only when the time-space logic state machine determines that the current process has reached the preset process end state, a write instruction is sent to the digital ledger module. The write instruction carries a timestamp and process completion evidence data. After receiving the write instruction, the digital ledger module binds and stores the timestamp, process completion evidence data and logical identity ID to form an immutable process quality traceability record, and synchronously updates the beam status in the digital twin system.
[0077] The beam yard management method based on QR code technology provided in this embodiment can be applied to the beam yard management system based on QR code technology provided in any of the above embodiments and achieve the same technical effect, which will not be elaborated here.
[0078] Example 3 The embodiments of this application can be applied to Figure 2 The electronic device (computer device) shown may be, but is not limited to, mobile terminals such as mobile phones, tablets, handheld computers, and personal digital assistants (PDAs), smart home devices such as smart TVs and smart cameras, wearable devices such as smart bracelets, smartwatches, and smart glasses, or other desktop, laptop, notebook, ultra-mobile personal computer (UMPC), netbook, and smart screen computer devices.
[0079] like Figure 2As shown, the electronic device 200 may include one or more of the following components: a processor 201, a memory 203, a communication interface 202, and a communication bus 204. The memory 203 can be connected to the processor 201 via the bus 204. The bus can transfer data between the processor 201 and the memory 203. The bus can be divided into an address bus, a data bus, a control bus, etc.
[0080] Processor 201 may include one or more processing cores. Processor 201 can connect to various parts within the electronic device 200 using various interfaces and lines. It performs various functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in memory 203, and by calling data stored in memory 203. For example, processor 201 may include an application processor (AP), a modem processor, a CPU, a graphics processing unit (GPU), an image signal processor (ISP), a controller, a video codec, a digital signal processor (DSP), a field-programmable gate array (FPGA), a programmable logic array (PLA), and / or a neural network processing unit (NPU). The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the content to be displayed; the NPU implements artificial intelligence (AI) functions; and the modem handles wireless communication. Different processing units can be independent devices or integrated into one or more processors. For example, the multiple processing units shown above are all integrated into a single SoC, or the AP is a separate semiconductor chip, while other processing units are integrated into a single SoC. This application does not limit this to any particular type.
[0081] The memory 203 may include random access memory (RAM), read-only memory (ROM), or non-transitory computer-readable storage medium. The memory 203 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 203 may include a program storage area and a data storage area. The program storage area may store instructions for implementing an operating system or instructions for at least one function, such as a beam yard management method based on QR code technology. The data storage area may store data created based on the use of the electronic device 200, such as real-time video.
[0082] In addition, those skilled in the art will understand that the structure of the electronic device 200 shown in the above figures does not constitute a limitation on the electronic device 200. The electronic device may include more or fewer components than shown, or combine certain components, or have different component arrangements. For example, the electronic device 200 may also include components such as a microphone, speaker, radio frequency circuit, sensor, audio circuit, power supply, and Bluetooth module, which will not be described in detail here.
[0083] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.
Claims
1. A beam yard management system based on QR code technology, characterized in that, It includes a perception layer, a processing layer, and an execution layer: The perception layer includes a laser deep etching unit and a video acquisition unit. The laser deep etching unit etches a QR code onto the surface of the precast beam using a laser to identify the logical identity ID of the precast beam and uses physical shadow features generated by the etching depth. The video acquisition unit is arranged in the work areas corresponding to different processes in the beam yard to collect real-time video streams from the work areas. The processing layer includes an edge computing unit, which runs an authentication module and a spatiotemporal logic state machine. The authentication module is configured to: recognize the QR code based on the real-time video stream, parse it to obtain the logical identity ID, extract the physical shadow features, and verify the authenticity of the precast beam's identity by comparing the physical shadow features with preset physical identifier attributes. The spatiotemporal logic state machine is configured to: establish a corresponding process verification instance for the precast beam after successful authentication, and determine the state transition based on the interaction between the construction equipment and the precast beam in the video stream. The execution layer includes a digital ledger module; The digital ledger module is communicatively connected to the edge computing unit; it receives a write instruction from the edge computing unit and persistently records it only when the time-space logic state machine determines that the current process has reached the preset process end state; the write instruction carries a timestamp and process completion evidence data.
2. The system of claim 1, wherein, The process by which the identity verification module verifies the authenticity of the precast beam includes: The QR code region in the video frame is located and a sub-region of the image is extracted using an object detection algorithm. The image sub-region is decoded by calling the QR code decoding module to obtain the logical identity ID; The shadow feature analysis module is invoked to perform light and shadow gradient analysis on the image sub-region and extract the shadow feature vector generated by the etching depth. If decoding is successful and the shadow feature vector matches the preset physical characteristics, then the identity verification is deemed successful; if decoding is successful but the shadow feature vector does not match the preset physical characteristics, then there is a risk of identifier replacement and an alarm is generated.
3. The system of claim 1, wherein, The state transition determination of the spatiotemporal logic state machine includes spatial topology constraints and time window constraints; The spatial topology constraint is used to determine the relative positional relationship between the construction equipment and the precast beam. When the relative positional relationship meets the preset geometric alignment conditions, a state transition is triggered. The time window constraint sets a minimum duration threshold and a maximum duration threshold for each state transition path. If the duration of the current state is less than the minimum duration threshold or greater than the maximum duration threshold, the spatiotemporal logic state machine determines that the process is invalid and resets, and does not perform the state transition operation.
4. The system of claim 1, wherein, The process verification example of the spatiotemporal logic state machine includes the initial state, beam position locking state, equipment alignment state, process execution state and process final state connected in sequence. The initial state is configured such that the beam QR code is not recognized or the work area is empty; The beam position locking status is triggered by the identity verification module confirming that the precast beam has entered the work area; The equipment alignment status is triggered when the geometric center of the construction equipment coincides with the anchorage area of the beam and the duration is greater than a first preset time. The execution status of the process is triggered by the recognition that the construction equipment is in working condition and the duration meets the process standard. The final state of the process is triggered by the detection that the construction equipment has detached from the beam.
5. The system of claim 1, wherein, The edge computing unit also runs an anti-occlusion re-identification algorithm; If the QR code is obscured and cannot be recognized after identity verification, the anti-occlusion re-identification algorithm is configured to: extract irregular textures on the surface of the beam as temporary fingerprints and track the displacement of feature points of the temporary fingerprints; If the displacement of the feature point does not exceed the preset displacement threshold, the spatiotemporal logic state machine maintains the current process verification instance of the beam and continues to execute the determination of subsequent states.
6. In the system according to claim 1, after receiving the write instruction, the digital ledger module binds and stores the timestamp, process completion evidence data and logical identity ID to form an immutable process quality traceability record, and synchronously updates the beam status in the digital twin system.
7. A beam yard management method based on QR code technology, applied to the beam yard management system based on QR code technology as described in any one of claims 1 to 6, characterized in that, Includes the following steps: The steps for identification and data acquisition are as follows: The laser deep etching unit etches the QR code onto the surface of the precast beam to identify the logical identity ID of the precast beam, and based on the physical shadow features generated by the etching depth; Video acquisition units are set up in the work areas corresponding to different processes in the beam yard, and the video acquisition units collect real-time video streams of the work areas. The data processing steps are as follows: Based on the real-time video stream, the QR code is identified, the logical identity ID is parsed, and the physical shadow features are extracted. The authenticity of the precast beam's identity is verified by comparing the physical shadow features with the preset physical identifier attributes. After the identity verification is passed, a corresponding process verification instance is established for the precast beam, and the state transition is determined based on the interaction state between the construction equipment and the precast beam in the video stream. The data recording steps are as follows: When the time-space logic state machine determines that the current process has reached the preset process end state, it sends a write instruction to the digital ledger module. The write instruction carries a timestamp and process completion evidence data. After receiving the write instruction, the digital ledger module binds and stores the timestamp, process completion evidence data and logical identity ID to form an immutable process quality traceability record, and synchronously updates the beam status in the digital twin system.
8. A computer apparatus comprising a memory, a processor, and a computer program stored on the memory, wherein the computer program, when executed by the processor, causes the processor to perform the method of any one of claims 1 to 7. The processor executes the computer program to implement the steps of the method of claim 7.
9. A computer readable storage medium having stored thereon computer programs / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method of claim 7.
10. A computer program product comprising computer programs / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method of claim 7.