Material Detection and Monitoring Method and Device Based on UWB Positioning and Multi-View Linkage

Abstract

This invention discloses a material inspection and monitoring method and device based on UWB positioning and multi-view linkage. The method includes: acquiring the sample number of the target material, wherein the target material is equipped with a UWB tag, and the UWB tag and the sample number correspond one-to-one; collecting the inspection time of the target material entering each inspection area according to the UWB tag, and constructing a time-location correlation matrix; and combining the time-location correlation matrix with a preset multi-view relationship model with the inspection area to form multi-view monitoring content, thereby completing the monitoring of the target material inspection process. By constructing a time-location correlation matrix and piecing together multi-view monitoring content, the monitoring of the target material inspection process is completed, providing a basis for rapid and accurate traceability of the target material inspection process afterward, while improving the level of material quality supervision and control, promoting quality inspection efficiency and effectiveness, and realizing full-process tracking and monitoring of the material inspection process.

Description

Technical Field

[0001] This invention belongs to the field of monitoring technology, specifically relating to a material detection and monitoring method and device based on UWB positioning and multi-view linkage. Background Technology

[0002] With the continuous advancement of modern digital supply chain construction, higher requirements have been placed on the traceability, impartiality, and risk control of the material inspection process, which has also increased the importance attached to material quality supervision. Traditional material inspection relies on manual labor, which can no longer meet the needs for accuracy and efficiency in inspection business management and inspection data management.

[0003] The current testing process lacks full-process monitoring and tracking of material samples, making it impossible to conduct rapid and accurate traceability analysis afterward.

[0004] How to further improve the level of supervision and control of material quality, promote the improvement of quality testing efficiency, and realize the whole-process tracking and monitoring of material testing process are the problems that need to be solved at present. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a material inspection and monitoring method and device based on UWB positioning and multi-view linkage. The method includes: acquiring the sample number of the target material, wherein the target material is equipped with a UWB tag, and the UWB tag and sample number correspond one-to-one; collecting the inspection time of the target material entering each inspection area based on the UWB tag, and constructing a time-location correlation matrix; and stitching together multi-view monitoring content based on the time-location correlation matrix and a pre-set multi-view relationship model with the inspection area to complete the monitoring of the target material inspection process. The multi-view includes fixed and moving views. By constructing the time-location correlation matrix and stitching together multi-view monitoring content, the monitoring of the target material inspection process is completed, providing a foundation for rapid and accurate traceability of the target material inspection process afterward. Simultaneously, it improves the level of material quality supervision and control, promotes quality inspection efficiency, and achieves full-process tracking and monitoring of the material inspection process.

[0006] In a first aspect, the present invention provides a material detection and monitoring method based on UWB positioning and multi-view linkage, wherein the multi-view includes a fixed view and a moving view, and specifically includes the following steps:

[0007] Obtain the sample number of the target material. The target material is equipped with a UWB tag, and the UWB tag and the sample number correspond one-to-one.

[0008] Based on the UWB tags, the detection time of the target material entering each detection area is collected, and a time-location correlation matrix is ​​constructed.

[0009] Based on the time-location correlation matrix and combined with the preset multi-view relationship model with the detection area, multi-view monitoring content is spliced ​​together to complete the monitoring of the target material detection process.

[0010] Furthermore, the time-location correlation matrix includes multiple time-location data points, each of which includes the sample number, detection area number, detection time, detection location, event type, and task status.

[0011] Furthermore, the target materials are identified upon entering each testing area through the following steps:

[0012] Obtain the location of the target material and the coordinates of the vertices of the detection area;

[0013] Obtain the coordinates of the vertices of the region sequentially, and give the region edge vectors corresponding to the coordinates of two adjacent region vertices.

[0014] By combining the coordinates of any vertex in the region edge vector with the material location, we obtain the material location edge vector;

[0015] By analyzing the cross product of the edge vectors of each region and the corresponding material location edge vectors, the positional relationship between the target material and the detection area is determined, and the judgment of the target material entering the detection area is completed.

[0016] Furthermore, the cross product of each region's edge vector and the corresponding material location edge vector is specifically expressed as follows:

[0017]

[0018] in, The cross product of the vectors corresponding to the edge vectors of the i-th region. Let be the coordinates of the vertex of the i-th region. Let P be the edge vector of the region corresponding to the coordinates of the vertex of the i-th region, and let P be the location of the target material.

[0019] Furthermore, based on the time-location correlation matrix and combined with a pre-set multi-view relationship model with the detection area, multi-view monitoring content is pieced together to complete the monitoring of the target material detection process, specifically including:

[0020] The pre-defined multi-view relationship model with the detection area is fused with the time-location correlation matrix, and the time-location correlation matrix is ​​adjusted.

[0021] Based on the adjusted time-location correlation matrix, the monitoring time corresponding to the monitoring content of fixed viewpoint and / or moving viewpoint is obtained;

[0022] Based on the order of monitoring time, the monitoring content is spliced ​​together to complete the monitoring of the target material testing process.

[0023] Furthermore, the pre-defined multi-view relationship model with the detection area is fused with the temporal location correlation matrix, and the temporal location correlation matrix is ​​adjusted, specifically including:

[0024] Based on the correspondence between fixed viewpoint, moving viewpoint and detection area, and combined with the detection time in the time-location correlation matrix, the monitoring content captured by fixed viewpoint and moving viewpoint is numbered to obtain the video file number.

[0025] Based on the detection time and detection area corresponding to the video file number, the video file number is added to the corresponding time position data in the time position correlation matrix, thus completing the adjustment of the time position correlation matrix.

[0026] Furthermore, the setting of the relationship model between the fixed viewpoint and the detection area specifically includes the following steps:

[0027] Analyze the detection range of the detection area and the shooting range of the monitoring equipment with a fixed viewing angle to construct coverage constraints;

[0028] A location optimization model is established by combining the objective function for the number of devices and the coverage constraints;

[0029] Based on the location optimization model, a search tree is built based on the detection area and monitoring equipment. The tree is then solved by combining upper and lower bound strategies and pruning strategies to obtain the location setting results. This completes the location setting of fixed viewpoints in each detection area and provides a relationship model between fixed viewpoints and detection areas.

[0030] Furthermore, combining the objective function for the number of devices and the coverage constraints, a location optimization model is established, specifically including:

[0031] Obtain the monitoring location, orientation, viewing angle, and coverage radius of the monitoring equipment;

[0032] Based on the shooting range of the monitoring equipment, the monitoring location and coverage radius of the monitoring equipment are analyzed, the relationship between the monitoring distance and the coverage radius is established, and the coverage distance constraint is obtained. The monitoring distance is the distance between any point within the shooting range and the monitoring location.

[0033] Based on the shooting range of the monitoring equipment, the monitoring position, monitoring orientation and monitoring angle of the monitoring equipment are analyzed, the relationship between the monitoring angle and the monitoring orientation and monitoring angle is established, and the coverage angle constraint is given. Among them, the monitoring angle is the angle between any point within the shooting range and the straight line where the monitoring position is located.

[0034] By combining coverage distance and coverage angle conditions, the relationship between the detection range of each detection area and the shooting range of each monitoring device is established, and coverage constraints are given.

[0035] By integrating the objective function of minimizing the number of monitoring devices, a location optimization model is obtained.

[0036] Furthermore, based on the location optimization model, a search tree is constructed based on the detection area and monitoring equipment, and the solution is obtained by combining upper and lower bound strategies and pruning strategies, specifically including:

[0037] Traverse each detection area, select the node value of each monitoring device, and analyze the coverage constraints based on the selected node values;

[0038] If the coverage constraint is met, analyze the node values ​​of the monitoring devices in each detection area and give the lower bound of the detection area;

[0039] By searching the node values ​​of the monitoring devices in the detection area, a feasible solution can be obtained.

[0040] Analyze the relationship between feasible solutions and the current optimal solution, and update the current optimal solution;

[0041] If the new optimal solution is less than or equal to the upper bound of the detection region, update the upper bound of the detection region.

[0042] If the new optimal solution is greater than or equal to the upper bound of the detection region, prune the current detection region;

[0043] Repeat the update of the current optimal solution and the upper bound until convergence, and obtain the position setting results of each detection region.

[0044] Secondly, the present invention also provides a material detection and monitoring device based on UWB positioning and multi-view linkage, employing the material detection and monitoring method based on UWB positioning and multi-view linkage as described above, including:

[0045] The data acquisition module is used to acquire the sample number of the target material. The target material is equipped with a UWB tag, and the UWB tag and the sample number correspond one-to-one.

[0046] The matrix construction module is used to collect the detection time of target materials entering each detection area based on UWB tags and construct a time-location correlation matrix.

[0047] The monitoring splicing module is used to splice together monitoring content from multiple perspectives based on the time and location correlation matrix and the preset relationship model between multiple perspectives and the detection area, so as to complete the monitoring of the target material detection process. The multiple perspectives include fixed perspectives and moving perspectives.

[0048] The material detection and monitoring method and device based on UWB positioning and multi-view linkage provided by this invention has at least the following beneficial effects:

[0049] (1) By combining the correspondence between UWB tags and sample numbers with the detection time of the target material entering different detection areas, a time-location correlation matrix is ​​constructed, and monitoring content from fixed and / or moving perspectives is spliced ​​together based on this matrix to complete the full-process monitoring of the target material. This provides a basis for the rapid and accurate tracing of the detection process of the target material after the fact, while improving the level of material quality supervision and control, promoting the improvement of quality detection efficiency, and realizing the full-process tracking and monitoring of the material detection process.

[0050] (2) By integrating fixed and moving perspectives, full coverage of each detection area is achieved, providing a foundation for full-process tracking and monitoring of material detection procedures.

[0051] (3) By setting up electronic fences in each detection area, the movement of target materials in the detection process can be monitored in real time, and the changes in the position of target materials can be captured in a timely manner to improve detection efficiency. Attached Figure Description

[0052] Figure 1 A flowchart illustrating the material detection and monitoring method based on UWB positioning and multi-view linkage provided in this embodiment of the invention;

[0053] Figure 2 This is a schematic diagram of different detection areas provided in the embodiments of the present invention;

[0054] Figure 3 This is a flowchart illustrating the process of target materials entering different detection areas, as provided in embodiments of the present invention.

[0055] Figure 4 A flowchart illustrating the splicing of monitoring content from fixed and / or moving perspectives provided in an embodiment of the present invention;

[0056] Figure 5 This is a structural block diagram of a material detection and monitoring device based on UWB positioning and multi-view linkage provided in an embodiment of the present invention.

[0057] Among them, 201 is the data acquisition module; 202 is the matrix construction module; and 203 is the monitoring splicing module. Detailed Implementation

[0058] To better understand the above technical solutions, a detailed description of the solutions will be provided below in conjunction with the accompanying drawings and specific embodiments. Obviously, the described embodiments are merely some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0059] The terminology used in the embodiments of this invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. The singular forms “a,” “the,” and “the” as used in the embodiments of this invention and the appended claims are also intended to include the plural forms, and “multiple” generally includes at least two unless the context clearly indicates otherwise.

[0060] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that an article or device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such an article or device. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the article or device that includes said element.

[0061] In recent years, with the continuous advancement of modern digital supply chain construction, higher requirements have been placed on the traceability, impartiality, and risk control of the material testing process. The workload of material sampling inspection has been increasing year by year, and the pressure on material testing risk control and operational timeliness has also been increasing.

[0062] Currently, the following problems exist in the inspection of power equipment:

[0063] (1) The key processes such as the handover, wiring, testing, and data recording of material samples lack detailed video recordings;

[0064] (2) Traditional fixed cameras only cover a local area and cannot achieve seamless tracking of the entire detection process. They require manual switching of monitoring screens, which is inefficient and prone to missing key links.

[0065] (3) The location information of the material samples (such as RFID or QR code) is not linked to the video surveillance data in real time, which requires manual matching during tracing, which is time-consuming and prone to errors;

[0066] (4) Historical videos are stored in a scattered manner, making it difficult to quickly retrieve the entire video stream.

[0067] In existing technologies, Ultra-Wideband (UWB) transmits information by transmitting and receiving ultrashort pulse signals. These ultrashort pulse signals are extremely short in duration, reaching nanoseconds or even picoseconds. Due to the short duration of the pulse signals, they exhibit a very wide frequency band distribution in the frequency domain. For example, a signal with a center frequency of 4 GHz and a pulse width of 1 nanosecond will have a wide bandwidth, covering multiple frequency bands. UWB utilizes the orthogonality of signals to distinguish different signals. At the receiving end, correlation operations are performed using the same pulse template as at the transmitting end to extract useful information. A key advantage of UWB is its strong anti-interference capability; even in the presence of noise and other signal interference, as long as the pulse characteristics can be correctly identified, the original information can be accurately recovered.

[0068] UWB is a carrier-free communication technology. Instead of using traditional continuous carrier signals, it transmits data using nanosecond-level or even narrower pulses. This technology can operate over a wide frequency range, typically with a bandwidth greater than 500MHz, or a relative bandwidth (bandwidth to center frequency ratio) greater than 20%.

[0069] Currently, while UWB positioning can provide high-precision location information, it has not been deeply integrated with multi-view video, including fixed and mobile cameras, to form a transparent, end-to-end monitoring system that integrates time and space. Therefore, developing the capability for visualized, end-to-end control and automated data collection for material quality inspection is particularly crucial.

[0070] like Figure 1 As shown in the figure, this embodiment of the invention provides a material detection and monitoring method based on UWB positioning and multi-view linkage, the specific steps of which are as follows:

[0071] S101: Obtain the sample number of the target material.

[0072] The target materials are equipped with UWB tags, and each UWB tag corresponds to a sample number.

[0073] It is important to understand that, in order to achieve standardized monitoring of target materials during the testing process, multi-view monitoring is adopted in the testing area. This multi-view includes both fixed and moving views, as referenced... Figure 2High-definition monitoring PTZ cameras are deployed in each testing area (A1, A2, A3, A4, A5, A6). These cameras serve as fixed-viewpoint cameras to capture and monitor the sample receiving area, the waiting area, the testing area, and the tested area. They are responsible for 24-hour video recording and real-time storage to a hard disk recorder, ensuring that the materials and samples are under control throughout the entire process. The video storage time is no less than 6 months for subsequent traceability. Mobile cameras are provided to personnel receiving returned samples and conducting testing. These cameras serve as mobile perspectives, responsible for recording key operational details from a first-person perspective, segmented by scene and time period. The video is stored in real-time on a unified video platform storage server for no less than 6 months, ensuring that the testing process and data are clearly visible.

[0074] In the embodiments provided by this invention, following the principles of comprehensive coverage and minimal quantity, fixed-view monitoring devices are positioned within the detection area to ensure the continuity and integrity of monitoring. The relationship model between multiple perspectives and the detection area includes a fixed-view relationship model and a moving-view relationship model. The moving-view relationship model is determined based on actual detection needs. For a specific detection area, close-range, full-process monitoring is required. Moving-view monitoring devices are provided to personnel within the detection area, and a relationship is established between the moving viewpoint and the corresponding detection area, forming a moving-view relationship model. For example, moving-view monitoring devices are provided to personnel in the sample receiving area to monitor the entire process of materials within the receiving area. If there is no necessary monitoring requirement, fixed-view monitoring devices can also be used to monitor the material detection.

[0075] Setting up the relationship model between a fixed viewpoint and the detection area includes the following steps:

[0076] Analyze the detection range of the detection area and the shooting range of the monitoring equipment with a fixed viewing angle to construct coverage constraints;

[0077] A location optimization model is established by combining the objective function for the number of devices and the coverage constraints;

[0078] Based on the location optimization model, a search tree is constructed based on the detection area and monitoring equipment. This tree is then solved using upper and lower bound strategies and pruning strategies to obtain the location setting results. This completes the location setting for fixed viewpoints in each detection area, and a relationship model between fixed viewpoints and detection areas is presented. The search tree consists of multiple subtrees, each subtree contains multiple subtree nodes, each subtree corresponds to a detection area, and each subtree node corresponds to a monitoring device.

[0079] Furthermore, combining the objective function for the number of devices and the coverage constraints, a location optimization model is established, specifically including:

[0080] Obtain the monitoring location, orientation, viewing angle, and coverage radius of the monitoring equipment;

[0081] Based on the shooting range of the monitoring equipment, the monitoring location and coverage radius of the monitoring equipment are analyzed, the relationship between the monitoring distance and the coverage radius is established, and the coverage distance constraint is obtained. The monitoring distance is the distance between any point within the shooting range and the monitoring location.

[0082] Based on the shooting range of the monitoring equipment, the monitoring position, monitoring orientation and monitoring angle of the monitoring equipment are analyzed, the relationship between the monitoring angle and the monitoring orientation and monitoring angle is established, and the coverage angle constraint is given. Among them, the monitoring angle is the angle between any point within the shooting range and the straight line where the monitoring position is located.

[0083] By combining coverage distance and coverage angle conditions, the relationship between the detection range of each detection area and the shooting range of each monitoring device is established, and coverage constraints are given.

[0084] By integrating the objective function of minimizing the number of monitoring devices, a location optimization model is obtained.

[0085] It's important to understand that the monitoring area of ​​each surveillance device can be represented as a sector, the monitored area of ​​which is determined by the monitoring location of the device. , monitoring orientation , monitoring perspective and coverage radius Decide.

[0086] For the i-th monitoring device, its covered sector area can be expressed as:

[0087]

[0088] The coverage distance constraint is specifically expressed as follows:

[0089]

[0090] Coverage distance constraints are specifically expressed as follows:

[0091]

[0092] Therefore, the coverage constraint is that the set of detection ranges of each detection area is within the set of monitoring ranges of each monitoring device, specifically expressed as:

[0093]

[0094] Where M represents the number of detection areas, and N represents the number of candidate monitoring device locations. For the j-th detection region, Let be the decision variable for the location of the i-th candidate monitoring device, where 1 indicates installation and 0 indicates no installation.

[0095] The objective function for the number of devices is specifically expressed as:

[0096]

[0097] The location optimization model is specifically represented as follows:

[0098]

[0099] in, Let J represent the set of locations of monitoring devices that can cover the j-th detection area.

[0100] In multiple detection areas, the optimal layout of monitoring equipment is determined based on the principle of comprehensive coverage and the use of the fewest monitoring devices. By establishing the objective function for coverage constraints and the number of devices, the location optimization model is completed.

[0101] The next step is to solve the location optimization model. The specific solution process is as follows:

[0102] First, a search tree needs to be built. The search tree consists of multiple subtrees, each subtree contains multiple subtree nodes, each subtree corresponds to a detection area, and each subtree node corresponds to a monitoring device. Initially, all monitoring devices are not selected (i.e., all...). At this point, we check if the coverage constraint is met. Clearly, the initial solution does not meet the coverage constraint because there is no monitoring equipment to cover any detection area.

[0103] Furthermore, based on the location optimization model, a search tree is constructed based on the detection area and monitoring equipment, and the solution is obtained by combining upper and lower bound strategies and pruning strategies, specifically including:

[0104] Traverse each detection area, select the node value of each monitoring device, and analyze the coverage constraints based on the selected node values;

[0105] If the coverage constraint is met, analyze the node values ​​of the monitoring devices in each detection area and give the lower bound of the detection area;

[0106] By searching the node values ​​of the monitoring devices in the detection area, a feasible solution can be obtained.

[0107] Analyze the relationship between feasible solutions and the current optimal solution, and update the current optimal solution;

[0108] If the new optimal solution is less than or equal to the upper bound of the detection region, update the upper bound of the detection region.

[0109] If the new optimal solution is greater than or equal to the upper bound of the detection region, prune the current detection region;

[0110] Repeat the update of the current optimal solution and the upper bound until convergence, and obtain the position setting results of each detection region.

[0111] First, select any uncovered detection area, and then select a monitoring device that can cover the detection area. Each detection area includes multiple monitoring devices, and each monitoring device selects a corresponding detection area to cover. For example, if detection area j can be covered by monitoring devices i1, i2, or i3, then three subtree nodes are branched from the current node, and each subtree node selects one of the monitoring devices. For each detection area, it is checked again whether there are any uncovered detection areas. If so, the branching continues, selecting another uncovered detection area and choosing a monitoring device for it, until all detection areas are covered.

[0112] In the embodiments provided by this invention, during the analysis of each detection area, a lower bound for each detection area needs to be calculated. The lower bound represents the minimum number of monitoring devices that can be found in that detection area. This lower bound can be estimated using a greedy heuristic or a relaxation problem. The lower bound is the sum of the number of monitoring devices required for the currently covered detection areas and the number of uncovered detection areas. For example, for the current partial solution (some monitoring devices have been selected), the remaining detection areas need to be covered, and assuming that each remaining detection area only requires one monitoring device to cover, the lower bound is the number of selected monitoring devices plus the number of remaining detection zones.

[0113] During the search in the search tree, when a feasible solution is found (i.e., a solution that covers all detection areas), the global upper bound is updated using the number of such solutions. Then, the objective function value of this solution (i.e., the number of monitoring devices) is compared with the current upper bound. If the objective function value of this solution is less than the current upper bound, the upper bound is updated.

[0114] Subsequently, when searching other nodes, if the lower bound of a certain detection region is greater than or equal to the current upper bound, the subtree of that node can be pruned, because continuing to search that subtree will not yield a better solution. If the solution corresponding to a subtree node is the same as or worse than that of a previously searched subtree node, that node is pruned to avoid duplicate searches.

[0115] During the search process, each time a new feasible solution is found, if the number of monitoring devices is less than the currently recorded optimal solution, the optimal solution is updated. When a node is completely searched (i.e., all its child nodes have been processed) or pruned, the algorithm backtracks to its parent node to continue searching other branches. When all nodes in the search tree have been processed (i.e., the entire tree has been searched), the algorithm terminates, and the recorded optimal solution at this point is the location setting result.

[0116] In other implementations, termination conditions such as maximum search time or maximum number of iterations can be set. When these conditions are met, the algorithm stops and returns the currently found optimal solution.

[0117] Following the principles of comprehensive coverage and minimal quantity, the location identification points are bound to the monitoring equipment in the corresponding detection areas to ensure the continuity and integrity of monitoring.

[0118] In a specific example, there are three detection areas A, B, and C. There are four candidate monitoring device locations with the following coverage areas: Location 1 covers A and B; Location 2 covers A and C; Location 3 covers B and C; Location 4 covers C. The process for selecting the minimum number of monitoring devices to cover the three detection areas is as follows:

[0119] Initialize the data, with the set of uncovered regions {A, B, C}. Select the first uncovered region A as the branch. Choose position 1 or 2 that can cover A. Create two subproblems:

[0120] Subproblem 1: Select position 1 (covering A and B), the uncovered area is {C};

[0121] Subproblem 2: Select position 2 (covering A and C), the uncovered area is {B}.

[0122] For subproblem 1: the lower bound is 1 (location 1 already selected) + 1 (at least one monitoring device is required to cover C) = 2. For subproblem 2: the lower bound is 1 (location 2 already selected) + 1 (at least one monitoring device is required to cover B) = 2.

[0123] Suppose that the current upper bound is initialized to infinity.

[0124] For subproblem 1, branch on the uncovered region C. Positions 2, 3, or 4 can be chosen to cover C. For example, choosing position 3 (covering B and C) results in a total of 2 monitoring devices (positions 1 and 3), covering all regions. The upper bound is then updated to 2. For subproblem 2, branch on the uncovered region B. Positions 1 or 3 can be chosen to cover B. For example, choosing position 3 (covering B and C) also results in a total of 2 monitoring devices. In subsequent searches, if a lower bound greater than or equal to 2 is encountered, prune the branch. Multiple solutions are found, with position combinations of (1,3), (2,3), or (1,2), etc., but a minimum of 2 monitoring devices are required to cover all regions.

[0125] By following the steps above, and combining the branch and bound method with coverage constraints, the problem of optimizing the layout of monitoring equipment can be effectively solved, ensuring that all areas are covered and the number of monitoring devices is minimized.

[0126] In the problem of optimizing the layout of surveillance equipment, a systematic search of all possible combinations of surveillance equipment ensures that no potential optimal solution is overlooked. This guarantees that, given the locations of candidate surveillance equipment and area coverage requirements, the precise layout scheme with the minimum number of surveillance equipment is found.

[0127] By estimating upper and lower bounds and employing feasibility pruning strategies, the search space can be effectively reduced, unnecessary computations avoided, and high-quality solutions found within limited computational resources and time. The correctness and reliability of the final monitoring equipment layout scheme can be ensured by backtracking the search process and checking the processing and pruning decisions for each branch.

[0128] S102: Based on the UWB tags, collect the detection time of the target material entering each detection area and construct a time-location correlation matrix.

[0129] The time-location correlation matrix includes multiple time-location data points, each of which includes the sample number, detection area number, detection time, detection location, event type, and task status.

[0130] In this example, a positioning system based on UWB tags is used to locate the target material's entry and exit times in the detection areas of each monitoring PTZ camera in real time, forming a correlation matrix between detection time and detection location (PTZ camera number), enabling real-time retrieval of monitoring content from a fixed viewing angle. It is understood that multiple UWB base stations are set up in each detection area. When a target material enters the detection area, it is assigned a corresponding UWB tag. Each UWB tag corresponds one-to-one with the sample number of the target material; that is, a unique UWB tag can be identified through the sample number, and a unique sample number can be identified through the UWB tag. When the target material is at any location in any of the detection areas, its current location can be determined by the UWB base station. This allows us to obtain information such as the detection area number, detection time, detection location, event type, and task status. The detection area number is a unique identifier for each detection area, and each detection area number corresponds to a monitoring PTZ camera with a fixed viewing angle. The detection time is the time corresponding to the determination of the current location. The detection location is the current location. The event type is either entering or leaving the detection area. The task status refers to the status corresponding to different detection areas. For example, the task status in the sample receiving area is sample receiving, and the task status in the detection area is detection.

[0131] By constructing a time-location correlation matrix, we can clearly understand the time sequence of target materials entering and leaving each detection area and their corresponding location trajectory, thereby providing strong data support for warehouse management. This enables us to promptly detect abnormal behaviors, such as target materials remaining in areas where they should not be for extended periods or frequently entering and leaving specific areas. Consequently, we can take corresponding measures, such as adjusting sample storage locations or strengthening zone management.

[0132] Furthermore, referring to Figure 3 Once the target materials enter each testing area, the following steps are used to determine their status:

[0133] Obtain the location of the target material and the coordinates of the vertices of the detection area;

[0134] Obtain the coordinates of the vertices of the region sequentially, and give the region edge vectors corresponding to the coordinates of two adjacent region vertices.

[0135] By combining the coordinates of any vertex in the region edge vector with the material location, we obtain the material location edge vector;

[0136] By analyzing the cross product of the edge vectors of each region and the corresponding material location edge vectors, the positional relationship between the target material and the detection area is determined, and the judgment of the target material entering the detection area is completed.

[0137] The cross product of the edge vectors of each region and the corresponding material location edge vectors is specifically expressed as:

[0138]

[0139] in, The cross product of the vectors corresponding to the edge vectors of the i-th region. Let be the coordinates of the vertex of the i-th region. Let P be the edge vector of the region corresponding to the coordinates of the vertex of the i-th region, and let P be the location of the target material.

[0140] Each detection area is equipped with a corresponding electronic fence. An electronic fence is a virtual boundary used in UWB positioning systems to restrict or monitor the movement range of target materials. When a target material enters or leaves this virtual boundary, the system triggers a corresponding operation. In the example provided in this invention, each detection area corresponds to one electronic fence. When a target material enters or leaves this detection area, corresponding fixed-view and / or moving-view devices need to be activated for monitoring. It can be understood that in this example, the electronic fence is used to monitor the movement of target materials during the detection process. For example, when a target material enters the detection area, its task status, event type, and detection time can be automatically recorded. When the target material reaches the designated detection area, a corresponding monitoring task can be triggered, achieving full-process monitoring of the target material detection process and improving detection efficiency.

[0141] In one specific implementation, the electronic fence corresponding to a certain detection area is an irregular quadrilateral, and the coordinates of the four vertices of the area are arranged in a counterclockwise order as follows: , The location of the supplies is Following a counter-clockwise order, the region edge vectors are obtained sequentially. .

[0142] For the region vertex coordinates The corresponding region edge vector is The edge vector of the material location is Calculate the cross product of the region edge vector and the material location edge vector. Similarly, calculate the cross product of the edge vectors for other regions. , If the cross product of all the edge vectors of the regions is greater than or equal to 0, then the material location P is inside the polygon; conversely, if there exists any cross product less than 0, then the material location P is outside the polygon.

[0143] In another specific implementation, the electronic fence corresponding to a certain detection area is an irregular quadrilateral, and the coordinates of the four vertices of the area are arranged in clockwise order as follows: , The location of the supplies is Following a clockwise order, the region edge vectors are obtained sequentially. .

[0144] For the region vertex coordinates The corresponding region edge vector is The edge vector of the material location is Calculate the cross product of the region edge vector and the material location edge vector. Similarly, calculate the cross product of the edge vectors for other regions. , If the cross product of all region edge vectors is less than or equal to 0, then the material location P is inside the polygon; conversely, if there exists any cross product greater than 0, then the material location P is outside the polygon.

[0145] S103: Based on the time-location correlation matrix and combined with the preset multi-view relationship model with the detection area, multi-view monitoring content is spliced ​​together to complete the monitoring of the target material detection process.

[0146] Reference Figure 4 Specifically, it includes:

[0147] The pre-defined multi-view relationship model with the detection area is fused with the time-location correlation matrix, and the time-location correlation matrix is ​​adjusted.

[0148] Based on the adjusted time-location correlation matrix, the monitoring time corresponding to the monitoring content of fixed viewpoint and / or moving viewpoint is obtained;

[0149] Based on the order of monitoring time, the monitoring content is spliced ​​together to complete the monitoring of the target material testing process.

[0150] Furthermore, the pre-defined multi-view relationship model with the detection area is fused with the temporal location correlation matrix, and the temporal location correlation matrix is ​​adjusted, specifically including:

[0151] Based on the correspondence between fixed viewpoint, moving viewpoint and detection area, and combined with the detection time in the time-location correlation matrix, the monitoring content captured by fixed viewpoint and moving viewpoint is numbered to obtain the video file number.

[0152] Based on the detection time and detection area corresponding to the video file number, the video file number is added to the corresponding time position data in the time position correlation matrix, thus completing the adjustment of the time position correlation matrix.

[0153] Understandably, on the inspection line, target materials move between different inspection areas. Each inspection area is equipped with an electronic fence and a monitoring PTZ camera. UWB tags record the event types of target materials entering and leaving each inspection area in real time and generate corresponding inspection times, resulting in a corresponding time-location correlation matrix. Based on the above time-location correlation matrix, the monitoring content captured by the fixed-view monitoring PTZ camera in the corresponding inspection area is added to the corresponding time-location data. The monitoring content is added to each time-location data through video file numbers, which consist of the sample number, inspection area number, event type, and inspection time. For example, for target material with sample number S01, when it enters inspection area number R01 at inspection time t1, the corresponding video file number is S01_R01_E_t1, where E indicates the event type is entry and L indicates the event type is exit.

[0154] Based on the time-location correlation matrix following the supplementary video file number, the monitoring time of each perspective is obtained. By splicing the monitoring content in the order of the monitoring time, the monitoring content of the target material throughout the entire process can be obtained, thus completing the full-process monitoring of the target material and facilitating the backtracking of the entire process of the target material inspection.

[0155] The material inspection and monitoring method based on UWB positioning and multi-view linkage provided in this invention adopts a technical approach of using a fixed PTZ camera to record the scene from a fixed perspective, a mobile camera to record details from a mobile perspective, and a UWB base station to record the trajectory. This enables real-time visual supervision and historical traceability of material sampling inspection videos, encompassing both the overall scene and key local close-ups. The fixed PTZ camera monitors the entire inspection process, achieving 24-hour video recording and real-time storage of the third-person perspective area, ensuring that samples are not replaced throughout the inspection process. The mobile camera monitors the details of the inspection operations, recording key operational details such as sample handover, wiring, testing, and data recording from a first-person perspective, ensuring transparency. The UWB indoor positioning technology is used to construct an automatic tracking algorithm for the inspected samples throughout the process, ensuring video linkage with corresponding area monitoring equipment and preventing technical and ethical risks throughout the inspection process.

[0156] By using an automatic video tracking algorithm for sample testing, the effectiveness of material quality supervision has been further improved. In the past, on-site spot checks and manual review of video recordings were carried out, which was labor-intensive and had low supervision coverage. Now, through the business transparency module, samples can be quickly retrieved with one click based on the sample task number. All testing tasks can be presented online in a "one-shot" manner, showing the entire process of sample reception, transportation, and testing, which significantly improves the efficiency of quality supervision.

[0157] Reference Figure 5 This invention provides a material detection and monitoring device based on UWB positioning and multi-view linkage, comprising:

[0158] The data acquisition module is used to acquire the sample number of the target material. The target material is equipped with a UWB tag, and the UWB tag and the sample number correspond one-to-one.

[0159] The matrix construction module is used to collect the detection time of target materials entering each detection area based on UWB tags and construct a time-location correlation matrix.

[0160] The monitoring splicing module is used to splice together monitoring content from multiple perspectives based on the time and location correlation matrix and the preset relationship model between multiple perspectives and the detection area, so as to complete the monitoring of the target material detection process. The multiple perspectives include fixed perspectives and moving perspectives.

[0161] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process of the described module can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0162] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention. Clearly, those skilled in the art can make various alterations and modifications to the invention without departing from its spirit and scope. Thus, if these modifications and variations of the invention fall within the scope of the claims and their equivalents, the invention is also intended to include these modifications and variations.

Claims

1. A material detection and monitoring method based on UWB positioning and multi-view linkage, characterized in that, Multi-view methods include fixed and moving viewpoints; material inspection and monitoring methods include: Obtain the sample number of the target material. The target material is equipped with a UWB tag, and the UWB tag and the sample number correspond one-to-one. Based on the UWB tags, the detection time of the target material entering each detection area is collected, and a time-location correlation matrix is ​​constructed. Based on the time-location correlation matrix and combined with the preset multi-view relationship model with the detection area, the monitoring content of the multi-view is spliced ​​to complete the monitoring of the target material detection process. The multi-view relationship model with the detection area includes the relationship model between the fixed view and the detection area and the relationship model between the moving view and the detection area. The setting of the relationship model between the fixed view and the detection area specifically includes: analyzing the detection range of the detection area and giving the shooting range of the monitoring equipment at the fixed view, and constructing coverage constraints. Obtain the monitoring location, orientation, viewing angle, and coverage radius of the monitoring equipment; Based on the shooting range of the monitoring equipment, the monitoring location and coverage radius of the monitoring equipment are analyzed, the relationship between the monitoring distance and the coverage radius is established, and the coverage distance constraint is obtained. The monitoring distance is the distance between any point within the shooting range and the monitoring location. Based on the shooting range of the monitoring equipment, the monitoring position, monitoring orientation and monitoring angle of the monitoring equipment are analyzed, the relationship between the monitoring angle and the monitoring orientation and monitoring angle is established, and the coverage angle constraint is given. Among them, the monitoring angle is the angle between any point within the shooting range and the straight line where the monitoring position is located. By combining coverage distance and coverage angle conditions, the relationship between the detection range of each detection area and the shooting range of each monitoring device is established, and coverage constraints are given. By integrating the objective function of minimizing the number of monitoring devices, a location optimization model is obtained; Traverse each detection area, select the node value of each monitoring device, and analyze the coverage constraints based on the selected node values; If the coverage constraint is met, analyze the node values ​​of the monitoring devices in each detection area and give the lower bound of the detection area; By searching the node values ​​of the monitoring devices in the detection area, a feasible solution can be obtained. Analyze the relationship between feasible solutions and the current optimal solution, and update the current optimal solution; If the new optimal solution is less than or equal to the upper bound of the detection region, update the upper bound of the detection region. If the new optimal solution is greater than or equal to the upper bound of the detection region, prune the current detection region; Repeatedly update the current optimal solution and the upper bound until convergence, to obtain the position setting results of each detection region, so as to complete the position setting of the fixed viewpoint in each detection region and give the relationship model between the fixed viewpoint and the detection region.

2. The material detection and monitoring method based on UWB positioning and multi-view linkage as described in claim 1, characterized in that, The time-location correlation matrix includes multiple time-location data points. Each time-location data point includes the sample number, detection area number, detection time, detection location, event type, and task status.

3. The material detection and monitoring method based on UWB positioning and multi-view linkage as described in claim 1, characterized in that, The target materials are identified upon entering each testing area through the following steps: Obtain the location of the target material and the coordinates of the vertices of the detection area; Obtain the coordinates of the vertices of the region sequentially, and give the region edge vectors corresponding to the coordinates of two adjacent region vertices. By combining the coordinates of any vertex in the region edge vector with the material location, we obtain the material location edge vector; By analyzing the cross product of the edge vectors of each region and the corresponding material location edge vectors, the positional relationship between the target material and the detection area is determined, and the judgment of the target material entering the detection area is completed.

4. The material detection and monitoring method based on UWB positioning and multi-view linkage as described in claim 3, characterized in that, The cross product of the edge vectors of each region and the corresponding material location edge vectors is specifically expressed as: ； Among them, C i V is the cross product of the vectors corresponding to the edge vectors of the i-th region. i Let V be the coordinates of the vertex of the i-th region. i V i+1 Let P be the edge vector of the region corresponding to the coordinates of the vertex of the i-th region, and let P be the location of the target material.

5. The material detection and monitoring method based on UWB positioning and multi-view linkage as described in claim 2, characterized in that, Based on the time-location correlation matrix and combined with the pre-set multi-view relationship model with the detection area, multi-view monitoring content is spliced ​​together to complete the monitoring of the target material detection process, specifically including: The pre-defined multi-view relationship model with the detection area is fused with the time-location correlation matrix, and the time-location correlation matrix is ​​adjusted. Based on the adjusted time-location correlation matrix, the monitoring time corresponding to the monitoring content of fixed viewpoint and / or moving viewpoint is obtained; Based on the order of monitoring time, the monitoring content is spliced ​​together to complete the monitoring of the target material testing process.

6. The material detection and monitoring method based on UWB positioning and multi-view linkage as described in claim 5, characterized in that, The pre-defined multi-view relationship model with the detection area is fused with the temporal location correlation matrix, and the temporal location correlation matrix is ​​adjusted, specifically including: Based on the correspondence between fixed viewpoint, moving viewpoint and detection area, and combined with the detection time in the time-location correlation matrix, the monitoring content captured by fixed viewpoint and moving viewpoint is numbered to obtain the video file number. Based on the detection time and detection area corresponding to the video file number, the video file number is added to the corresponding time position data in the time position correlation matrix, thus completing the adjustment of the time position correlation matrix.

7. A material detection and monitoring device based on UWB positioning and multi-view linkage, characterized in that, The material detection and monitoring method based on UWB positioning and multi-view linkage as described in any one of claims 1-6 includes: The data acquisition module is used to acquire the sample number of the target material. The target material is equipped with a UWB tag, and the UWB tag and the sample number correspond one-to-one. The matrix construction module is used to collect the detection time of target materials entering each detection area based on UWB tags and construct a time-location correlation matrix. The monitoring splicing module is used to splice together monitoring content from multiple perspectives based on the time and location correlation matrix and the preset relationship model between multiple perspectives and the detection area, so as to complete the monitoring of the target material detection process. The multiple perspectives include fixed perspectives and moving perspectives.

Images