Intelligent displacement detector, detection system and detection method
Through sub-pixel image processing and feature point recognition technology, combining historical and current image feature points, the problems of low accuracy and high cost of traditional bridge displacement detection are solved, and high-precision, low-cost and reliable bridge displacement detection are achieved. Adapt to different environmental conditions.
Patent Information
- Application Number
- CN202411019734.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-29
- Publication Date
- 2024-12-06
AI Technical Summary
Traditional bridge displacement detection has problems such as low accuracy, high cost, environmental sensitivity and susceptibility to interference, which affects the accuracy and reliability of the detection.
By acquiring the current real scene image and converting it into a subpixel image, combining the targets and real feature points in the historical image, image processing and feature point recognition technology are used to detect the bridge displacement, and high-precision displacement detection is achieved.
It improves the accuracy and reliability of displacement detection, reduces costs, and can monitor the bridge displacement in real time and continuously under different environmental conditions, detect abnormal displacements in a timely manner, and prevent potential risks.
Smart Images

Figure CN119090957A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of image recognition, and in particular relates to an intelligent displacement detector, a detection system and a detection method. Background Art
[0002] Bridge displacement detection is a very important task in bridge engineering, which can help engineers understand the health status and structural safety of the bridge. Bridge displacement detection usually includes parameters such as vertical displacement (settlement), horizontal displacement (lateral displacement) and rotation angle.
[0003] Bridge displacement detection is one of the important means of bridge monitoring, but there are some limitations in practical applications, which may affect the accuracy and reliability of detection. High-precision sensors are usually expensive and sensitive to environmental conditions (such as temperature, humidity, vibration, etc.), requiring regular calibration and maintenance. Sensors may be affected by electromagnetic interference, mechanical damage or other external factors, resulting in data distortion.
[0004] Therefore, how to achieve a low-cost and high-precision bridge displacement detection has become a technical problem that needs to be solved urgently. Summary of the invention
[0005] In view of this, the embodiments of the present invention provide an intelligent displacement detector, a detection system and a detection method to solve the technical problems of the limitations of traditional bridge displacement detection.
[0006] A first aspect of an embodiment of the present invention provides an intelligent displacement detection method, the intelligent displacement detection method comprising:
[0007] The first acquisition step: acquiring a current real scene image, and converting the current real scene image into a first sub-pixel image;
[0008] The second acquisition step: acquiring historical target feature points in the second sub-pixel image corresponding to the standard image and second object feature points corresponding to a plurality of pre-selected objects; the standard image refers to a historical real scene image collected under normal conditions;
[0009] The first recognition step: identifying the current target feature points corresponding to each of the multiple targets in the first sub-pixel image;
[0010] A second recognition step: recognizing first physical object feature points corresponding to a plurality of pre-selected physical objects in the first sub-pixel image;
[0011] Detection step: Detect whether the bridge body is displaced based on the distance relationship between the multiple current target feature points, the multiple first physical feature points, the multiple historical target feature points and the multiple second physical feature points.
[0012] Furthermore, the detection step comprises:
[0013] Detection and extraction step: extracting a first current target feature point located in a preset image area from among the plurality of current target feature points; the target corresponding to the first current target feature point is located at the end of the bridge;
[0014] A first detection calculation step: calculating a first distance between the first current target feature point and a plurality of the first physical object feature points;
[0015] Detection and judgment step: if the plurality of first distances are all less than the first threshold, then based on the preset order of the plurality of current target feature points, calculating the second distance between the current target feature point and the plurality of first physical object feature points;
[0016] The second detection calculation step: sequentially calculating the third distances between adjacent current target feature points;
[0017] A third detection calculation step: based on a preset order of a plurality of historical target feature points, calculating a fourth distance between the historical target feature points and a plurality of the second physical object feature points;
[0018] Fourth detection calculation step: sequentially calculating fifth distances between adjacent historical target feature points;
[0019] Detection sub-step: Detect whether the bridge body is displaced based on the numerical differences among the second distance, the third distance, the fourth distance and the fifth distance.
[0020] Furthermore, the detection sub-step includes:
[0021] The first detection sub-step: subtracting the second distance from the fourth distance one by one according to the preset order to obtain a first difference;
[0022] Second detection sub-step: subtracting the third distance from the fifth distance one by one in the preset order to obtain a second difference;
[0023] The third detection sub-step: if the plurality of the first differences are all smaller than the second threshold, and the plurality of the second differences are all smaller than the third threshold, it is determined that the bridge body has not been displaced;
[0024] Fourth detection sub-step: if the number of the plurality of first differences greater than the second threshold exceeds a first preset number, and the plurality of the second differences are all less than a third threshold, then it is determined that the bridge body is displaced;
[0025] The fifth detection sub-step: if the number of the plurality of second difference values greater than the third threshold value exceeds a second preset number, it is determined that the target is abnormal.
[0026] Furthermore, the second identification step includes:
[0027] A first identification and acquisition step: acquiring a target Gaussian mixture model corresponding to the second sub-pixel image;
[0028] A second identification and acquisition step: acquiring an initial rectangular area where the preselected object is located in the second sub-pixel image, and expanding the initial rectangular area into a target rectangular area according to a preset ratio;
[0029] Identification and interception step: intercepting a current image corresponding to a target rectangular area in the first sub-pixel image;
[0030] The first recognition and extraction step: extracting the image area corresponding to the preselected object in the current image according to the target Gaussian mixture model;
[0031] The second identification and extraction step: extracting the center point of the image area and taking the center point as the current target feature point.
[0032] Furthermore, before the first identification and acquisition step, the method further includes:
[0033] Identification and acquisition sub-step: acquiring a first image area selected by a user in the second sub-pixel image;
[0034] Identification and selection sub-step: taking the first image area as an initial foreground area, and taking the non-first image area in the second sub-pixel image as an initial background area;
[0035] The first identification calculation sub-step: calculating a first mean value, a first covariance matrix and first weight coefficients of a plurality of first Gaussian components of the initial foreground area;
[0036] Second identification calculation sub-step: calculating a second mean, a second covariance matrix and second weight coefficients of a plurality of second Gaussian components of the initial background area;
[0037] A first identification construction sub-step: constructing a first Gaussian mixture model based on the first mean, the first covariance matrix and the first weight coefficient;
[0038] Second identification construction sub-step: constructing a second Gaussian mixture model based on the second mean, the second covariance matrix and the second weight coefficient;
[0039] A third identification calculation sub-step: calculating a first probability that each pixel in the second sub-pixel image belongs to the foreground according to the first Gaussian mixture model;
[0040] Fourth identification calculation sub-step: calculating, according to the second Gaussian mixture model, a second probability that each pixel in the second sub-pixel image belongs to the background;
[0041] The fifth recognition calculation sub-step: calculating the similarity between adjacent pixels;
[0042] Identification and matching sub-step: Based on the first probability, the second probability and the similarity corresponding to each pixel point, a maximum flow / minimum cut algorithm is used to match the new foreground area and the new background area;
[0043] Identify loop sub-step: replace the initial foreground area and the initial background area with the new foreground area and the new background area, and iterate the step of calculating the first mean, the first covariance matrix and the first weight coefficient of the initial foreground area and subsequent steps until the maximum number of iterations is reached;
[0044] Identification and determination sub-step: taking the final second Gaussian mixture model as the target Gaussian mixture model.
[0045] Furthermore, the first identification and extraction step includes:
[0046] Calculation sub-step: Substituting the pixel value of each pixel in the current image into the target Gaussian mixture model to obtain a feature probability value output by the target Gaussian mixture model;
[0047] Selection sub-step: taking the pixel points whose feature probability values are greater than a preset value as the image area corresponding to the pre-selected real object.
[0048] Furthermore, before the first detection sub-step, the method further includes:
[0049] Detection calculation steps: Calculate the load of the bridge according to the number of vehicles passing through;
[0050] Detection matching step: matching the deformation error value corresponding to the load amount;
[0051] Detection acquisition step: obtain a preset error value, add the preset error value to the deformation error value, and obtain the second threshold value.
[0052] A second aspect of an embodiment of the present invention provides an intelligent displacement detection device, comprising:
[0053] A first acquisition unit, configured to acquire a current real scene image and convert the current real scene image into a first sub-pixel image;
[0054] A second acquisition unit is used to acquire historical target feature points in a second sub-pixel image corresponding to a standard image and second object feature points corresponding to a plurality of pre-selected objects; the standard image refers to a historical real scene image collected under normal conditions;
[0055] A first recognition unit, used to recognize current target feature points corresponding to each of the multiple targets in the first sub-pixel image;
[0056] A second recognition unit, used to recognize first physical feature points corresponding to a plurality of pre-selected physical objects in the first sub-pixel image;
[0057] The detection unit is used to detect whether the bridge body is displaced based on the distance relationship between the multiple current target feature points, the multiple first physical feature points, the multiple historical target feature points and the multiple second physical feature points.
[0058] A third aspect of an embodiment of the present invention provides an intelligent displacement detection system, the intelligent displacement detection system comprising a cloud server, an intelligent displacement detector and a terminal device;
[0059] The intelligent displacement detector is used to collect the current real scene image and send the current real scene image to the server;
[0060] The server is used to obtain a current real scene image and convert the current real scene image into a first sub-pixel image;
[0061] The server is used to obtain historical target feature points in a second sub-pixel image corresponding to a standard image and second object feature points corresponding to a plurality of pre-selected objects; the standard image refers to a historical real scene image collected under normal conditions;
[0062] The server is used to identify current target feature points corresponding to each of the multiple targets in the first sub-pixel image;
[0063] The server is used to identify first physical feature points corresponding to a plurality of pre-selected physical objects in the first sub-pixel image;
[0064] The server is used to detect whether the bridge body is displaced according to the distance relationship between the multiple current target feature points, the multiple first physical feature points, the multiple historical target feature points and the multiple second physical feature points;
[0065] The server is used to send the detection result to the terminal device.
[0066] A fourth aspect of an embodiment of the present invention provides an intelligent displacement detector, the device comprising: a memory, a processor, and an intelligent displacement detection program stored in the memory and executable on the processor, the intelligent displacement detection program being configured to implement an intelligent displacement detection method as described in any one of the above methods.
[0067] Compared with the prior art, the embodiments of the present invention have the following beneficial effects: by acquiring the current real scene image and converting it into a first sub-pixel image with higher precision, and combining the historical target feature points and physical feature points in the second sub-pixel image formed by the standard image, the subtle changes of the current target feature points and the physical feature points can be accurately identified. This sub-pixel level processing method greatly improves the accuracy of displacement detection and can detect the slight displacement of the bridge body. The distance relationship between multiple target feature points and multiple physical feature points is used to ensure the reliability and accuracy of target positioning. Through the joint analysis of multi-point features, the detection error caused by single-point error is reduced, and the stability and reliability of displacement detection are improved. This technical solution does not rely on expensive high-precision hardware equipment, but achieves high-precision displacement detection through image processing and feature point recognition technology. This method can be implemented using existing image acquisition equipment and computing resources, significantly reducing the overall cost of detection. By quickly acquiring and processing the current real scene image and comparing it with the historical image, the displacement of the bridge body can be monitored in real time and continuously. Abnormal displacement of the bridge body can be discovered in time to prevent potential risks. The method can adapt to image changes under different lighting and weather conditions, and by identifying physical feature points and target feature points, it ensures that the bridge displacement can be accurately detected under various environmental conditions. In summary, the bridge displacement detection method provided by the present invention not only realizes high-precision and low-cost bridge displacement detection, but also has the characteristics of strong reliability and good real-time performance. BRIEF DESCRIPTION OF THE DRAWINGS
[0068] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments or related technical descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0069] Figure 1 A schematic flow chart of an intelligent displacement detection method provided by the present invention is shown;
[0070] Figure 2 A schematic diagram of an intelligent displacement detection system provided by the present invention is shown;
[0071] Figure 3 A schematic diagram of an intelligent displacement detection device provided by an embodiment of the present invention is shown;
[0072] Figure 4 A schematic diagram of an intelligent displacement detector provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION
[0073] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present invention. However, it should be clear to those skilled in the art that the present invention may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present invention.
[0074] The embodiments of the present invention provide an intelligent displacement detector, a detection system and a detection method to solve the technical problem of the limitations of traditional bridge displacement detection.
[0075] First, the present invention provides an intelligent displacement detection system, which includes a cloud server, an intelligent displacement detector and a terminal device; the intelligent displacement detector is used to collect a current real scene image and send the current real scene image to the server; the server is used to obtain the current real scene image and convert the current real scene image into a first sub-pixel image; the server is used to obtain historical target feature points in a second sub-pixel image corresponding to a standard image and second physical feature points corresponding to multiple pre-selected physical objects; the standard image refers to a historical real scene image collected under a normal state; the server is used to identify the current target feature points corresponding to each of the multiple targets in the first sub-pixel image; the server is used to identify the first physical feature points corresponding to the multiple pre-selected physical objects in the first sub-pixel image; the server is used to detect whether the bridge body is displaced based on the distance relationship between the multiple current target feature points, the multiple first physical feature points, the multiple historical target feature points and the multiple second physical feature points; the server is used to send the detection result to the terminal device.
[0076] Secondly, the present invention provides an intelligent displacement detection method. Figure 1 , Figure 1 FIG. 2 shows a schematic flow chart of an intelligent displacement detection method provided by the present invention. Figure 1 As shown, the intelligent displacement detection method may include the following steps:
[0077] The first acquisition step: acquiring a current real scene image, and converting the current real scene image into a first sub-pixel image;
[0078] In order to improve the detection accuracy, the present invention converts the real scene image into a sub-pixel image. Wherein, the sub-pixel image refers to a technology in image processing and display technology that improves the details and quality of the image by refining the resolution of the image and dividing the pixels into smaller units (sub-pixels).
[0079] It is worth noting that in order to improve the recognition accuracy, the intelligent displacement detector and the target need to be installed in a specific position, see Figure 2 , Figure 2 FIG. 2 shows a schematic diagram of an intelligent displacement detection system provided by the present invention. Figure 2 As shown, the intelligent displacement detector is installed at both ends of the bridge body, and the target is installed on the structure of the bridge body. The structure of the bridge body includes but is not limited to piers, bridges and cable towers. Displacement detection needs to be performed separately for different bridge structures.
[0080] The second acquisition step: acquiring historical target feature points in the second sub-pixel image corresponding to the standard image and second object feature points corresponding to a plurality of pre-selected objects; the standard image refers to a historical real scene image collected under normal conditions;
[0081] The present invention determines whether the bridge body is displaced by comparing the position distribution between the target feature points and the physical feature points in the current real scene image and the historical real scene image. Therefore, it is necessary to obtain the historical target feature points in the second sub-pixel image corresponding to the standard image and the second physical feature points corresponding to multiple pre-selected physical objects to perform subsequent detection steps.
[0082] The first recognition step: identifying the current target feature points corresponding to each of the multiple targets in the first sub-pixel image;
[0083] The present invention has a target in a specific setting of the bridge body, and the current target feature point is obtained by identifying the code or specific shape on the target, wherein the current target feature point refers to the target center.
[0084] A second recognition step: recognizing first physical object feature points corresponding to a plurality of pre-selected physical objects in the first sub-pixel image;
[0085] Specifically, the second identification step specifically includes a first identification acquisition step, a second identification acquisition step, an identification interception step, a first identification extraction step, and a second identification extraction step:
[0086] A first identification and acquisition step: acquiring a target Gaussian mixture model corresponding to the second sub-pixel image;
[0087] The target Gaussian mixture model is a feature model trained based on the image features of the pre-selected object in the second sub-pixel image, and is used to screen feature points corresponding to the pre-selected object in the first sub-pixel image. The training logic of the target Gaussian mixture model is as follows:
[0088] Before the first identification and acquisition step, the method further includes the following steps:
[0089] Identification and acquisition sub-step: acquiring a first image area selected by a user in the second sub-pixel image;
[0090] Identification and selection sub-step: taking the first image area as an initial foreground area, and taking the non-first image area in the second sub-pixel image as an initial background area;
[0091] The first identification calculation sub-step: calculating a first mean value, a first covariance matrix and first weight coefficients of a plurality of first Gaussian components of the initial foreground area;
[0092] The clustering algorithm is used to calculate the clustering clusters of the pixel values in the initial foreground area, and the multiple clustering clusters are used as multiple first Gaussian components. The proportion of the pixel points corresponding to each first Gaussian component in all the pixel points is used as the first weight coefficient. According to the pixel values in the initial foreground area, the first mean and the first covariance matrix are calculated.
[0093] Second identification calculation sub-step: calculating a second mean, a second covariance matrix and second weight coefficients of a plurality of second Gaussian components of the initial background area;
[0094] The clustering algorithm is used to calculate the clustering clusters of the pixel values in the initial background area, and the multiple clustering clusters are used as multiple second Gaussian components. The proportion of the pixel points corresponding to each second Gaussian component in all the pixel points is used as the second weight coefficient. According to the pixel values in the initial foreground area, the second mean and the second covariance matrix are calculated.
[0095] A first identification construction sub-step: constructing a first Gaussian mixture model based on the first mean, the first covariance matrix and the first weight coefficient;
[0096] The pixel value of each pixel is regarded as a sample of a multi-dimensional Gaussian distribution. The first Gaussian mixture model is used to learn the pixel features in the initial foreground area. Among them, constructing a Gaussian mixture model is a conventional technology and will not be repeated here.
[0097] Second identification construction sub-step: constructing a second Gaussian mixture model based on the second mean, the second covariance matrix and the second weight coefficient;
[0098] The pixel value of each pixel is regarded as a sample of a multi-dimensional Gaussian distribution. The second Gaussian mixture model is used to learn the pixel features in the initial background area. Among them, constructing a Gaussian mixture model is a conventional technology and will not be repeated here.
[0099] A third identification calculation sub-step: calculating a first probability that each pixel in the second sub-pixel image belongs to the foreground according to the first Gaussian mixture model;
[0100] Fourth identification calculation sub-step: calculating, according to the second Gaussian mixture model, a second probability that each pixel in the second sub-pixel image belongs to the background;
[0101] The fifth recognition calculation sub-step: calculating the similarity between adjacent pixels;
[0102] Identification and matching sub-step: Based on the first probability, the second probability and the similarity corresponding to each pixel point, a maximum flow / minimum cut algorithm is used to match the new foreground area and the new background area;
[0103] An undirected graph is constructed according to the first probability, the second probability and the similarity corresponding to each pixel.
[0104] The principle of constructing an undirected graph is as follows: each pixel point is regarded as a pixel node in the undirected graph, and a foreground node and a background node are set in the undirected graph. The first probability is regarded as the edge between the pixel node and the foreground node, the second probability is regarded as the edge between the pixel node and the background node, and the similarity is regarded as the edge between different pixel nodes to obtain an undirected graph.
[0105] Among them, an undirected graph is a type of graph in which each edge has no direction, that is, the relationship between the two nodes connected by the edge is symmetrical. In an undirected graph, if there is an edge connecting node A and node B, then this edge can be expressed as (A, B), and this edge can also be expressed as (B, A), because the relationship connecting the two nodes is bidirectional. Nodes are the basic units in undirected graphs, also called vertices. Edges are line segments connecting nodes in undirected graphs, representing the relationship between nodes.
[0106] The maximum flow / minimum cut algorithm is used to optimize the image boundary that separates the foreground and the background. The maximum flow / minimum cut algorithm is used to perform graph cuts on the undirected graph to obtain a new foreground region and a new background region. The maximum flow / minimum cut algorithm is a traditional technology and will not be described in detail here.
[0107] Identify loop sub-step: replace the initial foreground area and the initial background area with the new foreground area and the new background area, and iterate the step of calculating the first mean, the first covariance matrix and the first weight coefficient of the initial foreground area and subsequent steps until the maximum number of iterations is reached;
[0108] Use the new foreground area and the new background area to re-estimate the parameters of the Gaussian mixture model. That is, update the mean, covariance and weight of the Gaussian mixture model. Through the above iterative steps, the Gaussian mixture model can better describe the color distribution of the foreground and background, thus providing a solid foundation for the subsequent image segmentation process.
[0109] Identification and determination sub-step: taking the final second Gaussian mixture model as the target Gaussian mixture model.
[0110] In this embodiment, the Gaussian mixture model constructed by using the mean, covariance matrix and weight coefficient of the initial foreground and background regions can more accurately describe the distribution of the foreground and background. Based on the statistical characteristics of the initial foreground and background regions, the Gaussian mixture model is quickly constructed to improve the efficiency of model training. Through iterative optimization, the optimal foreground and background regions are gradually approached, and the accuracy of segmentation is improved. By calculating the probability that each pixel belongs to the foreground and background respectively, and combining the similarity between adjacent pixels, the maximum flow / minimum cut algorithm is used to optimize the segmentation, so that the segmentation result is more accurate and detailed. The foreground and background regions are updated iteratively for multiple times, and the model parameters are gradually optimized to ensure that the final Gaussian mixture model can more accurately describe the foreground and background in the image. The technical solution continuously updates the foreground and background regions during each iteration, so that the model has a strong adaptability to different types of images. By limiting the maximum number of iterations, it is ensured that the calculation process is completed within a reasonable time, avoiding overfitting or waste of computing resources. In general, the technical solution significantly improves the segmentation effect and efficiency by optimizing the initial division, iterative update and probability calculation of the foreground and background regions.
[0111] A second identification and acquisition step: acquiring an initial rectangular area where the preselected object is located in the second sub-pixel image, and expanding the initial rectangular area into a target rectangular area according to a preset ratio;
[0112] Since the image area of the pre-selected object in the first sub-pixel image may be offset relative to the image area of the pre-selected object in the second sub-pixel image, it is necessary to expand the initial rectangular area where the pre-selected object is located to the target rectangular area according to a preset ratio to prevent the image area from being offset, resulting in the inability to select the image area of the pre-selected object in the first sub-pixel image.
[0113] Identification and interception step: intercepting a current image corresponding to a target rectangular area in the first sub-pixel image;
[0114] The first recognition and extraction step: extracting the image area corresponding to the preselected object in the current image according to the target Gaussian mixture model;
[0115] Specifically, the first recognition and extraction step includes a calculation sub-step and a selection sub-step:
[0116] Calculation sub-step: Substituting the pixel value of each pixel in the current image into the target Gaussian mixture model to obtain a feature probability value output by the target Gaussian mixture model;
[0117] Selection sub-step: taking the pixel points whose feature probability values are greater than a preset value as the image area corresponding to the pre-selected real object.
[0118] Since the target Gaussian mixture model is trained based on the pre-selected objects in the second sub-pixel image, it can be directly used to perform feature recognition on the pre-selected objects in the second sub-pixel image.
[0119] The second identification and extraction step: extracting the center point of the image area and taking the center point as the current target feature point.
[0120] In this embodiment, by substituting the pixel value of each pixel in the current image into the target Gaussian mixture model and calculating the feature probability value output by the model, the area in the image related to the pre-selected object can be effectively identified. Specifically, when the feature probability value of a certain pixel exceeds a preset threshold, the pixel is marked as belonging to the corresponding image area of the pre-selected object. The Gaussian mixture model can accurately describe the feature distribution of the target object, so that the identified image area has high accuracy and reliability.
[0121] Detection step: Detect whether the bridge body is displaced based on the distance relationship between the multiple current target feature points, the multiple first physical feature points, the multiple historical target feature points and the multiple second physical feature points.
[0122] Specifically, the detection step specifically includes a detection extraction step, a first detection calculation step, a detection judgment step, a second detection calculation step, a third detection calculation step, a fourth detection calculation step and a detection sub-step:
[0123] Detection and extraction step: extracting a first current target feature point located in a preset image area from among the plurality of current target feature points; the target corresponding to the first current target feature point is located at the end of the bridge;
[0124] Among them, the intelligent displacement detector and the first current target feature point are respectively set at different bridge ends. The position of the bridge end is relatively stable and not prone to displacement. Therefore, it is possible to judge whether the physical feature point has changed through the first current target feature point of the bridge end.
[0125] A first detection calculation step: calculating a first distance between the first current target feature point and a plurality of the first physical object feature points;
[0126] Since the feature points corresponding to the pre-selected physical object may be subject to object movement or image recognition anomalies, the present invention uses the first current target feature point, which is relatively fixed in position, to determine whether the feature points corresponding to the pre-selected physical object are abnormal, so as to improve the subsequent detection accuracy. That is, the feature points corresponding to the pre-selected physical object are checked to see whether they are abnormal by using the first current target feature point.
[0127] Detection and judgment step: if the plurality of first distances are all less than the first threshold, then based on the preset order of the plurality of current target feature points, calculating the second distance between the current target feature point and the plurality of first physical object feature points;
[0128] If the plurality of first distances are all smaller than the first threshold, the feature points corresponding to the pre-selected physical object are in a normal state, and thus subsequent detection steps can be performed.
[0129] The second detection calculation step: sequentially calculating the third distances between adjacent current target feature points;
[0130] A third detection calculation step: based on a preset order of a plurality of historical target feature points, calculating a fourth distance between the historical target feature points and a plurality of the second physical object feature points;
[0131] Fourth detection calculation step: sequentially calculating fifth distances between adjacent historical target feature points;
[0132] Detection sub-step: Detect whether the bridge body is displaced based on the numerical differences among the second distance, the third distance, the fourth distance and the fifth distance.
[0133] The second distance is the distance relationship between the current target feature point and the first physical feature point, the fourth distance is the distance relationship between the historical target feature point and the second physical feature point, the third distance is the distance relationship between adjacent current target feature points, and the fifth distance is the distance relationship between adjacent historical target feature points. By comparing the above distance relationships, it is detected whether the bridge body has been displaced. The specific logic is as follows:
[0134] Specifically, the detection sub-steps specifically include a first detection sub-step, a second detection sub-step, a third detection sub-step, a fourth detection sub-step and a fifth detection sub-step:
[0135] The first detection sub-step: subtracting the second distance from the fourth distance one by one according to the preset order to obtain a first difference;
[0136] The first difference is used to characterize the distance difference between the target feature point and the physical feature point in the current real scene image and the historical real scene image. The distance difference is used to evaluate the degree of displacement. The larger the first difference is, the greater the degree of displacement is.
[0137] Second detection sub-step: subtracting the third distance from the fifth distance one by one in the preset order to obtain a second difference;
[0138] The second difference is used to characterize the distance difference between the target feature points in the current real scene image and the historical real scene image. The distance difference is used to evaluate whether there is a target anomaly between the target feature points.
[0139] The third detection sub-step: if the plurality of the first differences are all smaller than the second threshold, and the plurality of the second differences are all smaller than the third threshold, it is determined that the bridge body has not been displaced;
[0140] Fourth detection sub-step: if the number of the plurality of first differences greater than the second threshold exceeds a first preset number, and the plurality of the second differences are all less than a third threshold, then it is determined that the bridge body is displaced;
[0141] The fifth detection sub-step: if the number of the plurality of second difference values greater than the third threshold value exceeds a second preset number, it is determined that the target is abnormal.
[0142] In this embodiment, the first current target feature points in the preset image area are extracted, and these targets are located at the bridge end, which ensures the accuracy and validity of the initial data. The distance between the first current target feature point and the first physical feature point is calculated, and a preliminary distance check is performed by comparing with the first threshold. If these distances are all less than the preset first threshold, the valid measurement data is screened out. On the basis of the preliminary check, based on the preset sorting of multiple current target feature points, the distance between the current target feature point and the first physical feature point, as well as the distance between adjacent current target feature points, are calculated in sequence. This sorting and comprehensive distance calculation method ensures the continuity and correlation of the measurement data and helps to identify small displacement changes. By calculating the distance between the historical target feature point and the second physical feature point, and comparing the distance between adjacent historical target feature points, a calibration reference based on historical data is provided. This comparison not only takes into account the change of the current state, but also combines the historical state to ensure the accuracy and reliability of the detection result. Then, by comparing the numerical differences between the second distance and the third distance of the current target feature point and the fourth distance and the fifth distance of the historical target feature point, it is possible to accurately detect whether the bridge body has been displaced. The analysis method of numerical differences can keenly capture subtle changes in the bridge structure and provide timely early warning information.
[0143] According to the preset order, the second distance related to the current target feature point is subtracted from the fourth distance related to the historical target feature point one by one to obtain the first difference. Similarly, the third distance between the current target feature points is subtracted from the fifth distance between the historical target feature points one by one to obtain the second difference. This difference calculation method can accurately capture the slight changes between the current state and the historical state. If multiple first differences are all less than the second threshold, and multiple second differences are all less than the third threshold, it can be determined that the bridge body has not been displaced, indicating the stability of the bridge structure. If the number of multiple first differences greater than the second threshold exceeds the first preset number, but multiple second differences are all less than the third threshold, it indicates that the bridge body has been displaced. At this time, by comparing the size and number of the difference, the existence of displacement can be accurately judged. If the number of multiple second differences greater than the third threshold exceeds the second preset number, the target is determined to be abnormal. This judgment can effectively identify abnormal situations in the measurement process and ensure the reliability of the detection results. This technical solution not only considers the changes in single data, but also provides a comprehensive assessment of the displacement state of the bridge body through multi-level difference calculation and comprehensive analysis. Multiple comparisons of differences and the judgment method of preset numbers increase the rigor of detection and avoid erroneous judgments caused by single data errors. Furthermore, through strict numerical difference analysis and threshold judgment mechanism, this technical solution greatly improves the reliability of bridge displacement detection.
[0144] As an optional embodiment of the present invention, before the first detection sub-step, the method further includes:
[0145] Detection calculation steps: Calculate the load of the bridge according to the number of vehicles passing through;
[0146] Since it is not necessary to calculate the exact load for the bridge displacement, it is only necessary to calculate the rough load based on the number of vehicles passing through. For example, the number of vehicles passing through is multiplied by the average vehicle weight to obtain the load of the bridge. The number of vehicles passing through refers to the number of vehicles currently on the bridge.
[0147] Since the load of the bridge body has a certain influence on the displacement of the bridge body, and the displacement caused by the load is a normal displacement, it is necessary to set a certain redundancy for the load displacement to prevent false detection and the like.
[0148] Detection matching step: matching the deformation error value corresponding to the load amount;
[0149] Detection acquisition step: obtain a preset error value, add the preset error value to the deformation error value, and obtain the second threshold value.
[0150] The preset error value is a value set according to the normal displacement range. The preset error value is added to the deformation error value to set a certain redundancy based on the preset error value.
[0151] In this embodiment, by acquiring the current real-scene image and converting it into a first sub-pixel image with higher precision, and combining the historical target feature points and physical feature points in the second sub-pixel image formed by the standard image, the subtle changes of the current target feature points and physical feature points can be accurately identified. This sub-pixel level processing method greatly improves the accuracy of displacement detection and can detect the slight displacement of the bridge body. The distance relationship between multiple target feature points and multiple physical feature points is used to ensure the reliability and accuracy of target positioning. Through the joint analysis of multi-point features, the detection error caused by single-point error is reduced, and the stability and reliability of displacement detection are improved. This technical solution does not rely on expensive high-precision hardware equipment, but achieves high-precision displacement detection through image processing and feature point recognition technology. This method can be implemented using existing image acquisition equipment and computing resources, significantly reducing the overall cost of detection. By quickly acquiring and processing the current real-scene image and comparing it with the historical image, the displacement of the bridge body can be monitored in real time and continuously. Abnormal displacement of the bridge body can be discovered in time to prevent potential risks. The method can adapt to image changes under different lighting and weather conditions, and by identifying physical feature points and target feature points, it ensures that the bridge displacement can be accurately detected under various environmental conditions. In summary, the bridge displacement detection method provided by the present invention not only realizes high-precision and low-cost bridge displacement detection, but also has the characteristics of strong reliability and good real-time performance.
[0152] like Figure 3 The present invention provides an intelligent displacement detection device 3, see Figure 3 , Figure 3 FIG. 4 shows a schematic diagram of an intelligent displacement detection device provided by the present invention. Figure 3 An intelligent displacement detection device is shown, comprising:
[0153] A first acquisition unit 31 is used to acquire a current real scene image and convert the current real scene image into a first sub-pixel image;
[0154] A second acquisition unit 32 is used to acquire historical target feature points in a second sub-pixel image corresponding to a standard image and second object feature points corresponding to a plurality of pre-selected objects; the standard image refers to a historical real scene image collected under normal conditions;
[0155] A first recognition unit 33, used to recognize current target feature points corresponding to each of the multiple targets in the first sub-pixel image;
[0156] A second recognition unit 34, configured to recognize first physical feature points corresponding to a plurality of pre-selected physical objects in the first sub-pixel image;
[0157] The detection unit 35 is used to detect whether the bridge body is displaced based on the distance relationship between the multiple current target feature points, the multiple first physical feature points, the multiple historical target feature points and the multiple second physical feature points.
[0158] The present invention provides an intelligent displacement detection device, which can accurately identify the subtle changes of the current target feature points and the physical feature points by acquiring the current real scene image and converting it into a first sub-pixel image with higher precision, and combining the historical target feature points and the physical feature points in the second sub-pixel image formed by the standard image. This sub-pixel level processing method greatly improves the accuracy of displacement detection and can detect the slight displacement of the bridge body. The distance relationship between multiple target feature points and multiple physical feature points is used to ensure the reliability and accuracy of target positioning. Through the joint analysis of multi-point features, the detection error caused by single-point error is reduced, and the stability and reliability of displacement detection are improved. This technical solution does not rely on expensive high-precision hardware equipment, but realizes high-precision displacement detection through image processing and feature point recognition technology. This method can be realized by using existing image acquisition equipment and computing resources, which significantly reduces the overall cost of detection. By quickly acquiring and processing the current real scene image and comparing it with the historical image, the displacement of the bridge body can be monitored in real time and continuously. The abnormal displacement of the bridge body can be discovered in time to prevent potential risks. The method can adapt to image changes under different lighting and weather conditions, and by identifying physical feature points and target feature points, it ensures that the bridge displacement can be accurately detected under various environmental conditions. In summary, the bridge displacement detection method provided by the present invention not only realizes high-precision and low-cost bridge displacement detection, but also has the characteristics of strong reliability and good real-time performance.
[0159] Figure 4 FIG. 1 is a schematic diagram of an intelligent displacement detector provided by an embodiment of the present invention. Figure 4 As shown, an intelligent displacement detector 4 of this embodiment includes: a processor 40, a memory 41, and a computer program 42 stored in the memory 41 and executable on the processor 40, such as an intelligent displacement detection program. When the processor 40 executes the computer program 42, the steps in the above-mentioned intelligent displacement detection method embodiments are implemented, such as Figure 1 Alternatively, when the processor 40 executes the computer program 42, the functions of each unit in the above-mentioned device embodiments are realized, for example Figure 2 Function of the unit shown.
[0160] Exemplarily, the computer program 42 may be divided into one or more units, which are stored in the memory 41 and executed by the processor 40 to complete the present invention. The one or more units may be a series of computer program instruction segments capable of completing specific functions, which are used to describe the execution process of the computer program 42 in the intelligent displacement detector 4. For example, the computer program 42 may be divided into the following specific functions of each unit:
[0161] A first acquisition unit, configured to acquire a current real scene image and convert the current real scene image into a first sub-pixel image;
[0162] A second acquisition unit is used to acquire historical target feature points in a second sub-pixel image corresponding to a standard image and second object feature points corresponding to a plurality of pre-selected objects; the standard image refers to a historical real scene image collected under normal conditions;
[0163] A first recognition unit, used to recognize current target feature points corresponding to each of the multiple targets in the first sub-pixel image;
[0164] A second recognition unit, used to recognize first physical feature points corresponding to a plurality of pre-selected physical objects in the first sub-pixel image;
[0165] The detection unit is used to detect whether the bridge body is displaced based on the distance relationship between the multiple current target feature points, the multiple first physical feature points, the multiple historical target feature points and the multiple second physical feature points.
[0166] The intelligent displacement detector includes but is not limited to a processor 40 and a memory 41. Those skilled in the art will appreciate that Figure 4 It is only an example of an intelligent displacement detector 4 and does not constitute a limitation of an intelligent displacement detector 4. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the intelligent displacement detector may also include input and output devices, network access devices, buses, etc.
[0167] The processor 40 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.
[0168] The memory 41 may be an internal storage unit of the intelligent displacement detector 4, such as a hard disk or memory of the intelligent displacement detector 4. The memory 41 may also be an external storage device of the intelligent displacement detector 4, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the intelligent displacement detector 4. Furthermore, the memory 41 may also include both the internal storage unit and the external storage device of the intelligent displacement detector 4. The memory 41 is used to store the computer program and other programs and data required by the roaming control device. The memory 41 may also be used to temporarily store data that has been output or is to be output.
[0169] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.
[0170] It should be noted that the information interaction, execution process, etc. between the above-mentioned devices / units are based on the same concept as the method embodiment of the present invention. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.
[0171] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In actual applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated into a processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.
[0172] An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the above-mentioned method embodiments can be implemented.
[0173] An embodiment of the present invention provides a computer program product. When the computer program product runs on a mobile terminal, the mobile terminal can implement the steps in the above-mentioned method embodiments when executing the computer program product.
[0174] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may at least include: any entity or device that can carry the computer program code to the camera device / intelligent displacement detector, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. For example, a USB flash drive, a mobile hard disk, a magnetic disk or an optical disk.
[0175] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.
[0176] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.
[0177] In the embodiments provided by the present invention, it should be understood that the disclosed devices / network equipment and methods can be implemented in other ways. For example, the device / network equipment embodiments described above are only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.
[0178] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed over multiple network units.
[0179] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.
[0180] It should also be understood that the term "and / or" used in the present description and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.
[0181] As used in the present specification and the appended claims, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to monitoring, depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is monitored" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is monitored" or "in response to monitoring [described condition or event]", depending on the context.
[0182] In addition, in the description of the present specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.
[0183] References to "one embodiment" or "some embodiments" etc. described in the present specification mean that one or more embodiments of the present invention include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.
[0184] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.
Claims
1. An intelligent displacement detection method, characterized in that: The intelligent displacement detection method comprises: The first acquisition step: acquiring a current real scene image, and converting the current real scene image into a first sub-pixel image; The second acquisition step: acquiring historical target feature points in the second sub-pixel image corresponding to the standard image and second object feature points corresponding to a plurality of pre-selected objects; the standard image refers to a historical real scene image collected under normal conditions; The first recognition step: identifying the current target feature points corresponding to each of the multiple targets in the first sub-pixel image; A second recognition step: recognizing first physical object feature points corresponding to a plurality of pre-selected physical objects in the first sub-pixel image; Detection step: Detect whether the bridge body is displaced based on the distance relationship between the multiple current target feature points, the multiple first physical feature points, the multiple historical target feature points and the multiple second physical feature points.
2. The intelligent displacement detection method according to claim 1, characterized in that: The detection step comprises: Detection and extraction step: extracting a first current target feature point located in a preset image area from among the plurality of current target feature points; the target corresponding to the first current target feature point is located at the end of the bridge; A first detection calculation step: calculating a first distance between the first current target feature point and a plurality of the first physical object feature points; Detection and judgment step: if the plurality of first distances are all less than the first threshold, then based on the preset order of the plurality of current target feature points, calculating the second distance between the current target feature point and the plurality of first physical object feature points; The second detection calculation step: sequentially calculating the third distances between adjacent current target feature points; A third detection calculation step: based on a preset order of a plurality of historical target feature points, calculating a fourth distance between the historical target feature points and a plurality of the second physical object feature points; Fourth detection calculation step: sequentially calculating fifth distances between adjacent historical target feature points; Detection sub-step: Detect whether the bridge body is displaced based on the numerical differences among the second distance, the third distance, the fourth distance and the fifth distance.
3. The intelligent displacement detection method according to claim 2, characterized in that: The detection sub-step comprises: The first detection sub-step: subtracting the second distance from the fourth distance one by one in a preset order to obtain a first difference; Second detection sub-step: subtracting the third distance from the fifth distance one by one in a preset order to obtain a second difference; The third detection sub-step: if the plurality of the first differences are all smaller than the second threshold, and the plurality of the second differences are all smaller than the third threshold, it is determined that the bridge body has not been displaced; Fourth detection sub-step: if the number of the plurality of first differences greater than the second threshold exceeds a first preset number, and the plurality of the second differences are all less than a third threshold, then it is determined that the bridge body is displaced; The fifth detection sub-step: if the number of the plurality of second difference values greater than the third threshold value exceeds a second preset number, it is determined that the target is abnormal.
4. The intelligent displacement detection method according to claim 1, characterized in that: The second identification step comprises: A first identification and acquisition step: acquiring a target Gaussian mixture model corresponding to the second sub-pixel image; A second identification and acquisition step: acquiring an initial rectangular area where the preselected object is located in the second sub-pixel image, and expanding the initial rectangular area into a target rectangular area according to a preset ratio; Identification and interception step: intercepting a current image corresponding to a target rectangular area in the first sub-pixel image; The first recognition and extraction step: extracting the image area corresponding to the preselected object in the current image according to the target Gaussian mixture model; The second identification and extraction step: extracting the center point of the image area and taking the center point as the current target feature point.
5. The intelligent displacement detection method according to claim 4, characterized in that: Before the first identification and acquisition step, the method further includes: Identification and acquisition sub-step: acquiring a first image area selected by a user in the second sub-pixel image; Identification and selection sub-step: taking the first image area as an initial foreground area, and taking the non-first image area in the second sub-pixel image as an initial background area; The first identification calculation sub-step: calculating a first mean value, a first covariance matrix and first weight coefficients of a plurality of first Gaussian components of the initial foreground area; Second identification calculation sub-step: calculating a second mean, a second covariance matrix and second weight coefficients of a plurality of second Gaussian components of the initial background area; A first identification construction sub-step: constructing a first Gaussian mixture model based on the first mean, the first covariance matrix and the first weight coefficient; Second identification construction sub-step: constructing a second Gaussian mixture model based on the second mean, the second covariance matrix and the second weight coefficient; A third identification calculation sub-step: calculating a first probability that each pixel in the second sub-pixel image belongs to the foreground according to the first Gaussian mixture model; Fourth identification calculation sub-step: calculating, according to the second Gaussian mixture model, a second probability that each pixel in the second sub-pixel image belongs to the background; The fifth recognition calculation sub-step: calculating the similarity between adjacent pixels; Identification and matching sub-step: Based on the first probability, the second probability and the similarity corresponding to each pixel point, a maximum flow / minimum cut algorithm is used to match the new foreground area and the new background area; Identify loop sub-step: replace the initial foreground area and the initial background area with the new foreground area and the new background area, and iterate the step of calculating the first mean, the first covariance matrix and the first weight coefficient of the initial foreground area and subsequent steps until the maximum number of iterations is reached; Identification and determination sub-step: taking the final second Gaussian mixture model as the target Gaussian mixture model.
6. The intelligent displacement detection method according to claim 4, characterized in that: The first identification and extraction step comprises: Calculation sub-step: Substituting the pixel value of each pixel in the current image into the target Gaussian mixture model to obtain a feature probability value output by the target Gaussian mixture model; Selection sub-step: taking the pixel points whose feature probability values are greater than a preset value as the image area corresponding to the pre-selected real object.
7. The intelligent displacement detection method according to claim 3, characterized in that: Before the first detection sub-step, the method further includes: Detection calculation steps: Calculate the load of the bridge according to the number of vehicles passing through; Detection matching step: matching the deformation error value corresponding to the load amount; Detection acquisition step: obtain a preset error value, add the preset error value to the deformation error value, and obtain the second threshold value.
8. An intelligent displacement detection device, characterized in that: The intelligent displacement detection device comprises: A first acquisition unit, configured to acquire a current real scene image and convert the current real scene image into a first sub-pixel image; A second acquisition unit is used to acquire historical target feature points in a second sub-pixel image corresponding to a standard image and second object feature points corresponding to a plurality of pre-selected objects; the standard image refers to a historical real scene image collected under normal conditions; A first recognition unit, used to recognize current target feature points corresponding to each of the multiple targets in the first sub-pixel image; A second recognition unit, used to recognize first physical feature points corresponding to a plurality of pre-selected physical objects in the first sub-pixel image; The detection unit is used to detect whether the bridge body is displaced based on the distance relationship between the multiple current target feature points, the multiple first physical feature points, the multiple historical target feature points and the multiple second physical feature points.
9. An intelligent displacement detector, characterized in that: The intelligent displacement detector comprises: a memory, a processor, and an intelligent displacement detection program stored in the memory and executable on the processor, wherein the intelligent displacement detection program is configured to implement the intelligent displacement detection method according to any one of claims 1 to 7.
10. An intelligent displacement detection system, characterized in that: The intelligent displacement detection system includes a cloud server, an intelligent displacement detector and a terminal device; The intelligent displacement detector is used to collect the current real scene image and send the current real scene image to the server; The server is used to obtain a current real scene image and convert the current real scene image into a first sub-pixel image; The server is used to obtain historical target feature points in a second sub-pixel image corresponding to a standard image and second object feature points corresponding to a plurality of pre-selected objects; the standard image refers to a historical real scene image collected under normal conditions; The server is used to identify current target feature points corresponding to each of the multiple targets in the first sub-pixel image; The server is used to identify first physical feature points corresponding to a plurality of pre-selected physical objects in the first sub-pixel image; The server is used to detect whether the bridge body is displaced according to the distance relationship between the multiple current target feature points, the multiple first physical feature points, the multiple historical target feature points and the multiple second physical feature points; The server is used to send the detection result to the terminal device.
