A safety visual monitoring method for traffic engineering construction equipment

Abstract

The present invention relates to the field of image analysis technology, and specifically to a method for visually monitoring the safety of traffic engineering construction equipment, including: obtaining each undetermined edge pixel point in the edge image of the traffic engineering construction equipment to be monitored, and then screening out each false deformation pixel point; determining the repair weight of each repair reference pixel point according to the position, false deformation degree and gray value of each false deformation pixel point and each corresponding repair reference pixel point; determining the repair position of each false deformation pixel point according to the repair weight and position of each repair pixel point, and then obtaining a repaired edge image; based on the repaired edge image, performing deformation monitoring on the traffic engineering construction equipment to be monitored to obtain a deformation monitoring result. The present invention improves the accuracy of visual monitoring of the safety of traffic engineering construction equipment by repairing the false deformed edges, and is suitable for the field of false edge repair.

Description

Technical Field

[0001] The present invention relates to the technical field of image analysis, and in particular to a method for visually monitoring safety of traffic engineering construction equipment. Background Art

[0002] Visual monitoring usually refers to real-time monitoring and analysis without direct contact with the target object through modern computer vision technology, sensors and artificial intelligence algorithms. This monitoring method is often used to evaluate and ensure construction safety, especially in high-risk construction environments. It can capture safety hazards in the construction process in real time, such as equipment failure, worker safety, and structural stability. Among them, external scaffolding is a temporary support structure commonly used in transportation engineering construction. Especially in bridges, elevated roads and other projects, external scaffolding is not only used to provide workers with a working platform to ensure the safety of construction personnel, but also supports various heavy objects and equipment that may exist during the construction process. The stability, safety and overall condition of the external scaffolding are crucial to the safe and smooth progress of the project. The deformation of the external scaffolding may cause structural instability and increase work safety risks. Therefore, it is very important to monitor the deformation degree of the external scaffolding during construction.

[0003] The existing construction external scaffolding deformation monitoring obtains the external scaffolding edge image through Canny edge detection, and then determines the curvature of each edge pixel to calculate the deformation degree. However, due to reasons such as slight camera shaking, some edges may produce virtual deformation, which can also be called false deformation. The curvature calculated based on the false deformed edge will affect the calculation accuracy of the external scaffolding deformation degree, further resulting in low accuracy of the construction external scaffolding deformation monitoring results. Summary of the invention

[0004] In order to solve the technical problem that the false deformation edge in the edge image of the external scaffolding causes low accuracy of the deformation monitoring result of the construction external scaffolding, the purpose of the present invention is to provide a safe visual monitoring method for traffic engineering construction equipment, and the technical scheme adopted is as follows:

[0005] In a first aspect, an embodiment of the present invention provides a method for visually monitoring safety of traffic engineering construction equipment, the method comprising the following steps:

[0006] Acquire an edge image of the traffic engineering construction equipment to be monitored, and determine each to-be-determined edge pixel point in the edge image;

[0007] Analyze the local deformation of the edge according to the position of each undetermined edge pixel point, determine the false deformation degree of each undetermined edge pixel point, and select each false deformation pixel point based on the false deformation degree;

[0008] Determine a number of repair reference pixel points corresponding to each false deformed pixel point, wherein the repair reference pixel points refer to a first preset number of to-be-determined edge pixel points located around the false deformed pixel point and on the same edge as the false deformed pixel point;

[0009] Determine the restoration weight of each restoration reference pixel corresponding to each false deformed pixel according to the position of each false deformed pixel and each restoration reference pixel corresponding thereto, the false deformation degree and the gray value of each restoration reference pixel;

[0010] According to the restoration weight and position of each restoration pixel point corresponding to each false deformation pixel point, the restoration position of each false deformation pixel point is determined, thereby obtaining a restoration edge image;

[0011] Based on the repaired edge image, deformation monitoring is performed on the traffic engineering construction equipment to be monitored to obtain a deformation monitoring result.

[0012] In the above scheme, the false edges in the edge image are repaired to overcome the virtual deformation of some edges due to slight shaking of the camera. The deformation monitoring of the external scaffolding under construction is carried out based on the repaired edge image, which can effectively improve the accuracy of the deformation monitoring results.

[0013] In combination with the first aspect, in some possible implementations, determining each undetermined edge pixel point in the edge image includes:

[0014] For any edge pixel point in the edge image, count the number of edge pixel points adjacent to the edge pixel point in the eight-neighborhood direction of the edge pixel point;

[0015] If the number of edge pixels is greater than a preset pixel number threshold, the edge pixel is discarded; if the number of edge pixels is not greater than the preset pixel number threshold, the tilt angle change value of the edge pixel is determined according to the position of the edge pixel and the edge pixels located around the edge pixel;

[0016] If the tilt angle change value of the edge pixel point is greater than the angle change threshold, the edge pixel point is discarded; otherwise, the edge pixel point is used as a pending edge pixel point.

[0017] In the above scheme, when determining each pending edge pixel point, the number of edge pixels adjacent to the edge pixel point in the eight-neighborhood direction of the edge pixel point is used to initially determine whether to discard the edge pixel point. If the conditions are not met, the edge pixel point is again determined based on the change value of the inclination angle. Combining different conditions to analyze whether to discard the edge pixel point helps to improve the reference value of the selected pending edge pixel point and improve the rigor of the selection of the pending edge pixel point.

[0018] In combination with the first aspect, in some possible implementations, determining the tilt angle change value of the edge pixel point according to the positions of the edge pixel point and edge pixel points located around the edge pixel point includes:

[0019] Selecting the same number of edge pixels on both sides of the edge to which the edge pixel belongs as neighborhood pixels, taking the neighborhood pixels on one side of the edge pixel as first neighborhood pixels, and taking the neighborhood pixels on the other side of the edge pixel as second neighborhood pixels;

[0020] Determine the slope of the line between the edge pixel point and each first neighborhood pixel point as the first slope, and determine the slope of the line between the edge pixel point and each second neighborhood pixel point as the second slope;

[0021] The tilt angle change value of the edge pixel point is determined according to the difference between all the first slopes and all the second slopes corresponding to the edge pixel point.

[0022] In combination with the first aspect, in some possible implementations, determining the tilt angle change value of the edge pixel point according to the difference between all first slopes and all second slopes corresponding to the edge pixel point includes:

[0023] Converting an average value of all first slopes corresponding to the edge pixel point into an angle value, which is recorded as a first angle value;

[0024] Converting the average value of all second slopes corresponding to the edge pixel point into an angle value, which is recorded as a second angle value;

[0025] The difference between the first angle value and the second angle value is calculated and determined as the tilt angle change value of the edge pixel point.

[0026] In combination with the first aspect, in some possible implementations, analyzing the local deformation of the edge according to the position of each undetermined edge pixel point to determine the false deformation degree of each undetermined edge pixel point includes:

[0027] Taking any undetermined edge pixel point as the target point, selecting a second preset number of undetermined edge pixel points closest to the target point on the edge to which the target point belongs as adjacent pixel points of the target point;

[0028] According to the slope of the line between the target point and each of its adjacent pixels, the variance is used to analyze the significant change of the slope of the edge and determine the degree of false deformation of the target point.

[0029] In the above scheme, on the edge to which the target point belongs, the adjacent pixel points of the target point are selected in order of distance from small to large. By analyzing the stability of the slope distribution between the target point and each adjacent pixel point, the slope of the selected adjacent pixel points is significantly changed, which helps to improve the numerical accuracy of the false deformation degree. The false deformation degree can be used to determine whether each pending edge pixel point is a false edge pixel point, providing data support for the subsequent selection of false deformation pixel points.

[0030] In combination with the first aspect, in some possible implementations, selecting each false deformation pixel point based on the false deformation degree includes:

[0031] A false deformation threshold is set, and for all pending edge pixels, the pending pixels whose false deformation degree is greater than the false deformation threshold are regarded as false deformation pixels.

[0032] In combination with the first aspect, in some possible implementations, determining the restoration weight of each restoration reference pixel corresponding to each false deformed pixel according to the position of each false deformed pixel and each restoration reference pixel corresponding thereto, the false deformation degree and the grayscale value of each restoration reference pixel, includes:

[0033] For any false deformed pixel point, according to the positions of the false deformed pixel point and the corresponding repair reference pixel points, the distance between the false deformed pixel point and the corresponding repair reference pixel points is determined, and a negatively correlated normalization process is performed on the distance, which is used as a first repair weight factor of the corresponding repair reference pixel point;

[0034] The restoration weight of each restoration reference pixel point corresponding to the false deformed pixel point is determined according to the first restoration weight factor, the false deformation degree and the gray value of each restoration reference pixel point corresponding to the false deformed pixel point.

[0035] In combination with the first aspect, in some possible implementations, determining the restoration weight of each restoration reference pixel corresponding to the false deformed pixel according to the first restoration weight factor, the false deformation degree, and the grayscale value of each restoration reference pixel corresponding to the false deformed pixel includes:

[0036] For any restoration reference pixel point, negatively correlated normalization processing is performed on the false deformation degree of the restoration reference pixel point to obtain a second restoration weight factor;

[0037] Calculating the difference between the gray value of the repair reference pixel and the mode of the gray value on the edge to which the false deformed pixel belongs, performing negative correlation normalization processing on the difference to obtain a third repair weight factor;

[0038] Fusing the first restoration weight factor, the second restoration weight factor, and the third restoration weight factor of the restoration reference pixel to obtain a fused value;

[0039] The proportion of the fused value in the accumulated value of the fused values ​​of all the repair reference pixels corresponding to the false deformed pixel is calculated to determine the repair weight of the repair reference pixel.

[0040] In the above scheme, the greater the false deformation degree of the repair reference pixel, the less important the repair reference pixel is in the position interpolation analysis of the false edge pixel, so it is necessary to perform negative correlation processing on the false deformation degree; the grayscale difference between the real deformed pixel and its surrounding edge pixel is small, while the grayscale difference between the false edge pixel and the surrounding edge pixel is large, so the grayscale difference value is negatively correlated; the repair weight that integrates the first repair weight factor, the second repair weight factor and the third repair weight factor can effectively improve the accuracy of the deformation monitoring results.

[0041] In combination with the first aspect, in some possible implementations, determining the restoration position of each false deformed pixel point according to the restoration weight and position of each restoration pixel point corresponding to each false deformed pixel point includes:

[0042] For any false deformed pixel point, a weighted summation process is performed according to the restoration weight and position of each restoration pixel point corresponding to the false deformed pixel point to obtain the restoration position of the false deformed pixel point.

[0043] In the above scheme, for each false deformed pixel point generated by false deformation, in order to overcome the defect that the position of the false deformed pixel point is offset due to false deformation, the repair weight and position of each repair pixel point corresponding to the false deformed pixel point are combined to perform weighted summation processing, so as to achieve more reliable interpolation analysis and determine the repair position of the false deformed pixel point.

[0044] In combination with the first aspect above, in some possible implementations, performing deformation monitoring on the traffic engineering construction equipment to be monitored to obtain a deformation monitoring result based on the repaired edge image includes:

[0045] Performing curve fitting according to each edge pixel point on each edge in the repaired edge image to obtain each edge fitting curve;

[0046] According to the curvature of each data point on each edge fitting curve, it is determined whether the traffic engineering construction equipment to be monitored is deformed.

[0047] In the above scheme, each edge fitting curve is obtained by performing curve fitting on each edge pixel point on each edge in the repaired edge image. Compared with the original edge image, the authenticity of each edge in the repaired edge image is higher, which is beneficial to improve the numerical accuracy of subsequent curvature calculation. The accuracy-based curvature can enhance the accuracy of the deformation monitoring results of the construction external scaffolding.

[0048] In a second aspect, a traffic engineering construction equipment safety visual monitoring device is provided, the device comprising:

[0049] The first determination module is used to obtain an edge image of the traffic engineering construction equipment to be monitored and determine each to-be-determined edge pixel point in the edge image.

[0050] The second determination module is used to analyze the local deformation of the edge according to the position of each pending edge pixel point, determine the false deformation degree of each pending edge pixel point, and select each false deformed pixel point based on the false deformation degree; determine a number of repair reference pixel points corresponding to each false deformed pixel point, wherein the repair reference pixel points refer to a first preset number of pending edge pixel points located around the false deformed pixel point and on the same edge as the false deformed pixel point;

[0051] A third determination module is used to determine the restoration weight of each restoration reference pixel corresponding to each false deformed pixel according to the position of each false deformed pixel and each restoration reference pixel corresponding thereto, the false deformation degree and the gray value of each restoration reference pixel;

[0052] A fourth determination module is used to determine the restoration position of each false deformed pixel point according to the restoration weight and position of each restoration pixel point corresponding to each false deformed pixel point, thereby obtaining a restoration edge image;

[0053] The monitoring module is used to perform deformation monitoring on the traffic engineering construction equipment to be monitored based on the repaired edge image to obtain a deformation monitoring result.

[0054] In a third aspect, a server is provided, comprising a memory and a processor. The memory is used to store executable program code, and the processor is used to call and run the executable program code from the memory, so that the device executes the method in the first aspect or any possible implementation of the first aspect.

[0055] In a fourth aspect, a computer program product is provided, comprising: a computer program code, which, when executed on a computer, enables the computer to execute the method in the first aspect or any possible implementation of the first aspect.

[0056] In a fifth aspect, a computer-readable storage medium is provided, which stores a computer program code. When the computer program code runs on a computer, the computer executes the method in the above-mentioned first aspect or any possible implementation manner of the first aspect.

[0057] The present invention has the following beneficial effects:

[0058] The present invention provides a method for visual safety monitoring of traffic engineering construction equipment, which involves edge detection technology and can be applied to the field of false edge repair. Compared with the existing method of directly monitoring deformation based on the edge image of traffic engineering construction equipment, the present invention repairs the false edges in the edge image, overcomes the virtual deformation of some edges due to slight camera shaking and other reasons, and monitors the deformation of traffic engineering construction equipment based on the repaired edge image, which can effectively improve the accuracy of the deformation monitoring results. Determining each pending edge pixel point can avoid the pixel points at the intersection position in the edge image from affecting the accuracy of the curvature calculation result. At the same time, it reduces the calculation objects to a certain extent, which is conducive to improving the calculation efficiency; when analyzing the repair weight of the repair reference pixel point, not only the positional relationship between the false deformed pixel point and the repair reference pixel point is taken into account, but also the influence of the false deformation degree and gray value of the repair reference pixel point on the position repair result is taken into account, which helps to improve the accuracy of the repair weight of the repair reference pixel point and provides reliable data support for the subsequent determination of the repair position. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. 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 paying creative work.

[0060] Figure 1 A flow chart of a method for visually monitoring safety of traffic engineering construction equipment provided by an embodiment of the present invention;

[0061] Figure 2 is a grayscale image of the traffic engineering construction equipment to be monitored in an embodiment of the present invention;

[0062] Figure 3 is an edge image of the traffic engineering construction equipment to be monitored in an embodiment of the present invention;

[0063] Figure 4 is a flowchart of the steps of determining the false deformation degree of a target point in an embodiment of the present invention;

[0064] Figure 5A flowchart of the steps of determining the restoration weight of each restoration reference pixel corresponding to a false deformed pixel in an embodiment of the present invention;

[0065] Figure 6 The restored edge image of the traffic engineering construction equipment to be monitored in the embodiment of the present invention;

[0066] Figure 7 A schematic diagram of the structure of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0067] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects of the technical solutions proposed by the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0068] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0069] The application scenarios targeted by the present invention may be:

[0070] When photographing the construction scaffolding, some edges in the construction scaffolding image may be falsely deformed due to factors such as camera shaking and interference from debris. False deformation will affect the accuracy of the construction scaffolding deformation monitoring results. In order to overcome the impact of false deformation, the edges corresponding to the false deformation are repaired, and then the construction scaffolding deformation analysis is performed based on the repaired edges.

[0071] In order to overcome the adverse effects of false deformation edges on the deformation analysis of the external scaffolding under construction and improve the accuracy of the deformation monitoring results of the external scaffolding under construction, specifically, this embodiment provides a safe visual monitoring method for traffic engineering construction equipment, such as Figure 1 As shown, the following steps are included:

[0072] S1, obtaining an edge image of the traffic engineering construction equipment to be monitored, and determining each to-be-determined edge pixel point in the edge image.

[0073] The first step is to obtain the edge image of the traffic engineering construction equipment to be monitored.

[0074] In the first sub-step, a parallel light source is used to shoot the traffic engineering construction equipment to be monitored, and an image of the traffic engineering construction equipment to be monitored is obtained. In order to analyze the extracted image features, the image of the traffic engineering construction equipment to be monitored needs to be gray-scale processed, and a gray-scale image of the traffic engineering construction equipment to be monitored can be obtained. Figure 2 As shown, the grayscale processing can be implemented by weighted averaging method.

[0075] In the second sub-step, in order to eliminate the influence of illumination to a certain extent, the grayscale image of the traffic engineering construction equipment to be monitored is enhanced to obtain the grayscale image after image enhancement. The image enhancement can be implemented by histogram equalization. When performing deformation analysis, the object of analysis is edge deformation. Therefore, it is necessary to perform edge processing on the grayscale image after image enhancement to obtain the edge image of the traffic engineering construction equipment to be monitored. The edge image of the traffic engineering construction equipment to be monitored is as follows: Figure 3 As shown, the implementation method of edge processing can be Canny edge detection.

[0076] Among them, the implementation processes of weighted averaging method, histogram equalization processing and Canny edge detection are all existing technologies, which are not within the protection scope of the present invention and will not be elaborated in detail here.

[0077] The second step is to determine each pending edge pixel point in the edge image of the traffic engineering construction equipment to be monitored.

[0078] It should be noted that because the pixel points at the junction of two or more frames in the construction external scaffolding are easy to affect the accuracy of the curvature calculation results, in order to improve the accuracy of the curvature calculation results, the pixel points at the junction of the frames are eliminated, that is, disconnected, so that only each frame can be monitored for deformation in the future.

[0079] In this embodiment, taking any edge pixel in the edge image as an example, determining whether the edge pixel is a pending edge pixel may include:

[0080] In the first sub-step, the number of edge pixels adjacent to the edge pixel in the eight-neighborhood direction of the edge pixel is counted, and if the number of edge pixels is greater than a preset pixel number threshold, the edge pixel is discarded.

[0081] In this embodiment, the preset pixel number threshold is set to 2. If the number of edge pixels adjacent to the edge pixel in different directions is greater than 2, it means that the pixel is an intersection of three or more edges. The edge pixel can be directly removed without subsequent curvature calculation.

[0082] In the second sub-step, if the number of edge pixels is not greater than a preset pixel number threshold, a tilt angle change value of the edge pixel is determined according to the position of the edge pixel and edge pixels surrounding the edge pixel.

[0083] It should be noted that if the number of edge pixels adjacent to the edge pixel point in different directions is equal to 2, it means that the pixel point is the intersection of edges on both sides. The inclination angle change value of the edge pixel point can be quantified based on the difference between the average inclination angles on the two sides of the edge pixel point, so as to facilitate subsequent judgment on whether to discard the edge pixel point.

[0084] The step of determining the tilt angle change value of the edge pixel point may include:

[0085] First, an equal number of edge pixels are selected on both sides of the edge to which the edge pixel belongs as neighborhood pixels, the neighborhood pixels on one side of the edge pixel are taken as first neighborhood pixels, and the neighborhood pixels on the other side of the edge pixel are taken as second neighborhood pixels.

[0086] In this embodiment, in order to ensure that the number of edge pixel points selected on both sides remains consistent, the number of neighborhood pixel points selected on both sides cannot exceed the minimum number of edge pixel points on both sides. For example, if there are 5 and 10 edge pixels on both sides of the edge to which a certain edge pixel point belongs, then 5 adjacent edge pixel points are selected on both sides of the edge to which the edge pixel point belongs as neighborhood pixel points.

[0087] Secondly, the slope of the line between the edge pixel point and each first neighborhood pixel point is determined as the first slope, and the slope of the line between the edge pixel point and each second neighborhood pixel point is determined as the second slope.

[0088] In this embodiment, the edge pixel point is connected to each first neighborhood pixel point respectively to obtain each connecting line corresponding to the edge pixel point, and the slope of each connecting line is calculated as the first slope. Similarly, each second slope corresponding to the edge pixel point can be obtained.

[0089] Next, the tilt angle change value of the edge pixel point is determined according to the difference between all the first slopes and all the second slopes corresponding to the edge pixel point.

[0090] Specifically, the average value of all first slopes corresponding to the edge pixel point is converted into an angle value, recorded as the first angle value; the average value of all second slopes corresponding to the edge pixel point is converted into an angle value, recorded as the second angle value; the difference between the first angle value and the second angle value is calculated and determined as the inclination angle change value of the edge pixel point.

[0091] As an example, the calculation formula for the tilt angle change value of the edge pixel point can be:

[0092] ; In the formula, represents the change value of the inclination angle of the edge pixel point, arctan represents the inverse tangent function, v represents the first slope number and the second slope number corresponding to the edge pixel point, i represents the first slope sequence number and the second slope sequence number corresponding to the edge pixel point, Indicates the i-th first slope corresponding to the edge pixel point, represents the first angle value, Indicates the i-th second slope corresponding to the edge pixel point, represents the first angle value, Represents the absolute value symbol.

[0093] In the calculation formula of the tilt angle change value, the edge pixel point and the edge pixels on both sides of it are used to determine the average slope, and then the inverse tangent function is used to quantify the angle value of the average slope, and the difference in the angle values ​​on both sides is used as the tilt angle change on both sides of the edge pixel point; the larger the tilt angle change value, the greater the possibility that the edge pixel point belongs to the actual intersection point, and the smaller the tilt angle change value, the smaller the possibility that the edge pixel point belongs to the actual intersection point.

[0094] It is worth mentioning that in order to capture all the information of the scaffolding outside the construction site, there is no edge parallel to the X-axis in the image obtained based on the shooting angle. Therefore, when the X-axis change is divided by the Y-axis change, there will be no situation where the slope does not exist. In other words, the situation where there is no slope when the angle value approaches 90 degrees can be avoided.

[0095] In the third sub-step, if the tilt angle change value of the edge pixel point is greater than the angle change threshold, the edge pixel point is discarded; otherwise, the edge pixel point is used as a pending edge pixel point.

[0096] In this embodiment, the angle change threshold is set to 80°. If the tilt angle change value of the edge pixel is greater than 80°, it means that the edge pixel belongs to an intersection that should be removed and needs to be discarded. Otherwise, the edge pixel is used as a pending edge pixel. The angle change threshold can be set by the implementer according to the actual situation and is not specifically limited.

[0097] It should be noted that compared with the existing intersection detection algorithm, the implementation process of the pending edge pixel points in this embodiment can detect all intersections more accurately. At the same time, it is also beneficial to avoid misjudging false deformed pixels as intersections that need to be discarded, and can obtain pending edge pixel points with higher reference value, providing data support for subsequent false deformed pixel point screening.

[0098] So far, this embodiment has obtained each undetermined edge pixel point in the edge image of the traffic engineering construction equipment to be monitored for analyzing the false edge situation.

[0099] S2, analyzing the local deformation of the edge according to the position of each pending edge pixel point, determining the false deformation degree of each pending edge pixel point, and selecting each false deformation pixel point based on the false deformation degree.

[0100] It should be noted that after removing the pixels at the intersection position, it is necessary to screen out the edge pixels that need to be repaired, that is, the false deformed pixels, from multiple pending edge pixels. The false deformed pixels are determined to facilitate subsequent position repair processing and obtain an edge image that avoids the influence of false edges to a certain extent.

[0101] In the first step, the local deformation of the edge is analyzed according to the position of each pending edge pixel point, and the false deformation degree of each pending edge pixel point is determined.

[0102] It should be noted that false deformation is local deformation. If false deformation exists, the local deformation is large, so the slope of the local edge formed by the pending edge pixel and its surrounding pending edge pixels may change significantly. Therefore, the degree of false deformation of the pending edge pixel can be quantified by analyzing the local deformation corresponding to the pending edge pixel and its surrounding pending edge pixels.

[0103] In this embodiment, the calculation process of the false deformation degree of each undetermined edge pixel point is consistent. In order to reduce unnecessary description and facilitate understanding of the calculation process of the false deformation degree, any undetermined edge pixel point is taken as a target point, and the false deformation degree of the target point is determined by taking the target point as an example. Figure 4 As shown, the specific implementation process may include:

[0104] In the first sub-step, a second preset number of to-be-determined edge pixel points which are closest to the target point are selected on the edge to which the target point belongs as adjacent pixel points of the target point.

[0105] In this embodiment, the second preset number can be set to 4. The size of the second preset number can be set by the implementer according to the actual situation without specific limitation. The target point can correspond to 4 adjacent pixel points.

[0106] In another example, the step of determining adjacent pixels of the target point may include:

[0107] Calculate the distance between each pending edge pixel point on the edge to which the target point belongs and the target point, sort each pending edge pixel point in order of distance from small to large, obtain a sequence, and select the first second preset number of pending edge pixel points in the sequence as adjacent pixel points of the target point.

[0108] It should be noted that the multiple nearest pixel points involved in this embodiment all refer to multiple pixel points selected in sequence from small to large in the distance sequence, and will not be repeated later.

[0109] In the second sub-step, the variance is used to analyze the significant change of the slope of the edge based on the slope of the line between the target point and each of its adjacent pixels, so as to determine the degree of false deformation of the target point.

[0110] As an example, the calculation formula for the false deformation degree of the target point can be:

[0111] ; In the formula, Indicates the degree of false deformation of the target point, represents the target point, f represents the normalization function, R represents the number of adjacent pixels of the target point, p represents the sequence number of the adjacent pixels of the target point, represents the slope of the line between the target point and the pth adjacent pixel point of the target point, arctan represents the inverse tangent function, Represents the average value of the slope of the line between the target point and all its adjacent pixels. Represents the absolute value symbol.

[0112] In the calculation formula of the false deformation degree, in order to facilitate the subsequent threshold judgment of the false deformation degree and filter out the false deformation pixels, normalization processing is required. The normalization function can be implemented by maximum and minimum value normalization processing. Of course, normalization processing can also be implemented by other means; It can represent the degree of significant change in the angle obtained by the slope corresponding to the target point and the edge pixels around it. The greater the degree of significant change, the greater the possibility that the target point is an edge pixel caused by false deformation, so this value is positively correlated with the degree of false deformation; a positive correlation means that the dependent variable will increase as the independent variable increases, and the dependent variable will decrease as the independent variable decreases. The specific relationship can be a multiplication relationship, an addition relationship, or the power of an exponential function, which is determined by actual application.

[0113] In the second step, each false deformation pixel point is selected according to the false deformation degree of each pending edge pixel point.

[0114] In this embodiment, the greater the false deformation degree, the greater the possibility that the pending edge pixel is a false deformation pixel. A false deformation threshold is set, and for all pending edge pixels, the pending pixels whose false deformation degree is greater than the false deformation threshold are taken as false deformation pixels, thereby obtaining each false deformation pixel, which is essentially an edge pixel that needs to be repaired.

[0115] The false deformation threshold may be set to 0.2, and implementers may set the size of the false deformation threshold according to actual conditions without making specific limitations.

[0116] So far, this embodiment has obtained each false deformed pixel point in the edge image of the traffic engineering construction equipment to be monitored.

[0117] S3, determining a number of repair reference pixels corresponding to each false deformed pixel point; determining a repair weight of each repair reference pixel point corresponding to each false deformed pixel point according to the position of each false deformed pixel point and each repair reference pixel point corresponding to the false deformed pixel point, the false deformation degree and the gray value of each repair reference pixel point.

[0118] It should be noted that in order to repair the position of the false deformed pixel in the edge image, it is necessary to determine the weight of the edge pixels surrounding the false deformed pixel during the interpolation analysis, so as to determine the importance of the edge pixels surrounding the false deformed pixel in the repair process.

[0119] The first step is to determine a number of repair reference pixels corresponding to each false deformed pixel.

[0120] When repairing the position of the false deformed pixel point, it is necessary to refer to the position of the surrounding edge pixel points, so it is necessary to determine a plurality of corresponding repair reference pixel points for each false deformed pixel point. The repair reference pixel points refer to a first preset number of undetermined edge pixel points located around the false deformed pixel point and on the same edge as the false deformed pixel point. The first preset number can be set to 8. The implementer can set the number of repair reference pixel points according to the actual situation, without specific limitation.

[0121] In the second step, the restoration weight of each restoration reference pixel corresponding to each false deformed pixel is determined according to the position of each false deformed pixel and each restoration reference pixel corresponding thereto, the false deformation degree and the gray value of each restoration reference pixel.

[0122] In this embodiment, the calculation process of the restoration weight of each restoration reference pixel corresponding to each false deformed pixel is consistent. In order to reduce the necessary description, an arbitrary false deformed pixel is taken as an example to determine the restoration weight of each restoration reference pixel corresponding to the false deformed pixel. Figure 5 As shown, the specific implementation process may include:

[0123] In the first sub-step, according to the positions of the false deformed pixel points and their corresponding repair reference pixel points, the distance between the false deformed pixel points and their corresponding repair reference pixel points is determined, and a negatively correlated normalization processing is performed on the distance as the first repair weight factor of the corresponding repair reference pixel point.

[0124] In this embodiment, the first restoration weight factor is determined by the distance between the conventional false deformed pixel point and each corresponding restoration reference pixel point. Affected by factors such as noise points, the restoration achieved by the first restoration weight factor is not completely accurate, so it is necessary to determine the restoration weight based on the first restoration weight factor. The implementation process of determining the distance by two-point position is a prior art and is not within the scope of protection of the present invention, and will not be elaborated here.

[0125] As an example, the calculation formula of the first restoration weight factor for restoring the reference pixel point may be:

[0126] ; In the formula, Indicates the first restoration weight factor of the nth restoration reference pixel corresponding to the false deformed pixel, Represents the distance between the false deformed pixel and its corresponding nth repair reference pixel, represents the distance between the false deformed pixel and its corresponding lth repair reference pixel, l represents the sequence number of the repair reference pixel corresponding to the false deformed pixel, and m represents the number of repair reference pixels corresponding to the false deformed pixel.

[0127] In the calculation formula of the first restoration weight factor, the cumulative sum of the first restoration weight factors of each restoration reference pixel corresponding to the false deformed pixel is 1. It is a negative correlation. The larger the distance, the less important the position of the nth repair reference pixel is for the position repair of the false deformed pixel, and the smaller the first repair weight factor is.

[0128] In a second sub-step, a restoration weight of each restoration reference pixel corresponding to the false deformed pixel is determined according to a first restoration weight factor, a false deformation degree and a gray value of each restoration reference pixel corresponding to the false deformed pixel.

[0129] It should be noted that there is a situation where the first repair weight factor of the repair reference pixel is large, but the degree of false deformation is also large. The greater the degree of false deformation, the greater the possibility that the repair reference pixel is a false deformed pixel. The reliability of position repair using repair reference pixels that may be false deformed pixels is low. Therefore, the first repair weight factor can be corrected considering the degree of false deformation of each repair reference pixel.

[0130] The blur caused by false deformation causes the grayscale size of the same edge to change, while the grayscale size of the edge corresponding to normal deformation does not change. Therefore, the first repair weight factor can be corrected again according to the grayscale difference of each repair reference pixel. The repair weight obtained through two corrections is more accurate, which is conducive to improving the accuracy of subsequent determination of the repair position.

[0131] Specifically, for any repair reference pixel, a negatively correlated normalization process is performed on the false deformation degree of the repair reference pixel to obtain a second repair weight factor; the difference between the grayscale value of the repair reference pixel and the mode of the grayscale values ​​on the edge to which the false deformed pixel belongs is calculated, and the difference is negatively correlated normalization process is performed on the difference to obtain a third repair weight factor; the first repair weight factor, the second repair weight factor and the third repair weight factor of the repair reference pixel are fused to obtain a fused value; the proportion of the fused value in the cumulative value of the fused values ​​of all repair reference pixels corresponding to the false deformed pixel is calculated to determine the repair weight of the repair reference pixel.

[0132] As an example, the calculation formula for the restoration weight of the nth restoration reference pixel corresponding to the false deformed pixel may be:

[0133] ; In the formula, Indicates the restoration weight of the nth restoration reference pixel corresponding to the false deformed pixel, represents the first restoration weight factor of the nth restoration reference pixel corresponding to the false deformed pixel, exp represents an exponential function with a natural constant as the base, Indicates the false deformation degree of the nth repair reference pixel corresponding to the false deformation pixel. represents the second repair weight factor, f represents the normalization function, Indicates the gray value of the nth repair reference pixel corresponding to the false deformed pixel, Indicates the majority of gray values ​​on the edge to which the false deformed pixel belongs, represents the hyperparameter, represents the third repair weight factor, represents the absolute value sign, l and n represent the serial numbers of the repair reference pixels corresponding to the false deformed pixels, l can be equal to n, m represents the number of repair reference pixels corresponding to the false deformed pixels, represents the first restoration weight factor of the lth restoration reference pixel corresponding to the false deformed pixel, Indicates the false deformation degree of the lth repair reference pixel corresponding to the false deformation pixel, Indicates the gray value of the lth repair reference pixel corresponding to the false deformed pixel.

[0134] In the calculation formula of the repair weight, A negatively correlated normalized value that can characterize the degree of false deformation, The larger the value is, the greater the false deformation degree of the nth repair reference pixel is, and the smaller the importance of the nth repair reference pixel in participating in position repair should be, that is, The smaller; It can represent the difference between the gray value of the nth repair reference pixel and the gray value mode of the same edge. The larger the value is, the greater the change in the grayscale value of the nth repair reference pixel is, the greater the possibility that the nth repair reference pixel is a pixel generated by the false deformation, and the smaller the importance of the nth repair reference pixel in participating in position repair should be, that is, The smaller the repair weight, the greater the coordinate weight of the repair reference pixel position in the process of repairing the false deformed pixel position; the hyperparameter It can be used to avoid the situation where the denominator of the fraction is zero. The hyperparameter can be set to 0.1.

[0135] So far, this embodiment obtains the restoration weight of each restoration reference pixel corresponding to each false deformed pixel.

[0136] S4, determining the restoration position of each false deformed pixel point according to the restoration weight and position of each restoration pixel point corresponding to each false deformed pixel point, and then obtaining a restoration edge image.

[0137] In the first step, the restoration position of each false deformed pixel point is determined according to the restoration weight and position of each restoration pixel point corresponding to each false deformed pixel point.

[0138] In this embodiment, the repair position of each false deformed pixel is determined in the same manner. For ease of understanding, any false deformed pixel is taken as an example to determine the repair position. The specific implementation process may include:

[0139] Combining the concept of linear interpolation algorithm, weighted summation processing is performed according to the restoration weight and position of each restoration pixel corresponding to the false deformed pixel to obtain the restoration position of the false deformed pixel.

[0140] As an example, the calculation formula for the repair position of the false deformed pixel point can be:

[0141] ; In the formula, represents the repair position of the wth false deformed pixel, m represents the number of repair reference pixels corresponding to the false deformed pixel, n represents the sequence number of the repair reference pixels corresponding to the false deformed pixel, represents the restoration weight of the nth restoration reference pixel corresponding to the wth false deformation pixel, Indicates the position of the nth repair reference pixel corresponding to the wth false deformed pixel.

[0142] In the second step, the repaired edge image is obtained by repairing the position of each false deformed pixel.

[0143] In this embodiment, based on the repair position of each false deformed pixel point, refer to the calculation process of the false deformation degree in the first step S2 to determine the false deformation degree of each false deformed pixel point. If there are still false deformed pixels whose false deformation degree is greater than the false deformation threshold, interpolation repair is performed again based on the coordinate position of the neighborhood pixel points until the false deformation degree of all false deformed pixels is no greater than the false deformation threshold, and then the position of the false deformed pixel points in each edge is updated, so that the repaired edge image of the traffic engineering construction equipment to be monitored can be obtained. The repaired edge image of the traffic engineering construction equipment to be monitored is as follows: Figure 6 shown.

[0144] S5, based on the repaired edge image, deformation monitoring is performed on the traffic engineering construction equipment to be monitored to obtain a deformation monitoring result.

[0145] It should be noted that when analyzing the deformation monitoring results, the curvature change can be calculated based on all edge pixels in each edge of the repaired edge image to obtain the deformation degree of the traffic engineering construction equipment to be monitored. The specific implementation steps may include:

[0146] In the first step, curve fitting is performed according to each edge pixel point on each edge in the repaired edge image to obtain each edge fitting curve.

[0147] In this embodiment, in order to analyze the change of curvature, the least square method can be used to perform polynomial fitting on each edge pixel point on each edge to obtain each edge fitting curve. The implementation process of the least square method is prior art and is not within the scope of protection of the present invention, and will not be elaborated here.

[0148] In the second step, based on the curvature of each data point on each edge fitting curve, it is determined whether the construction equipment of the traffic engineering project to be monitored is deformed.

[0149] In this embodiment, the curvature of each data point on each edge fitting curve is determined, and the curvature threshold is determined in combination with the standard shape characteristics of each edge in the external construction scaffolding. If the curvature of more than three-quarters of the data points on any one or more edge fitting curves is greater than the curvature threshold, it can be determined that the monitored traffic engineering construction equipment is deformed; otherwise, it is determined that the monitored traffic engineering construction equipment is not deformed, and the deformation monitoring result is obtained.

[0150] It should be noted that the curvature threshold can be set to 3 degrees. In addition, since the deformation of the external construction scaffolding shows that the distribution of the single frame components is deformed as a whole, it is necessary to judge the external construction scaffolding with most of the curvatures greater than the curvature threshold as deformed. The implementer can set the proportion of data points greater than the curvature threshold in the entire edge fitting curve and the curvature threshold according to the actual situation, without specific limitation.

[0151] The embodiment of the present invention provides a method for visually monitoring the safety of traffic engineering construction equipment. The method first determines each edge of the traffic engineering construction equipment, and then selects edge pixels that need to be repaired, that is, false deformation pixels, from each edge; by setting different weights for each neighborhood pixel coordinate value, the repair position of the false deformation pixel is determined, thereby obtaining a repaired edge image; based on the repaired edge image, deformation monitoring is performed on the monitored traffic engineering construction equipment to obtain a deformation monitoring result. The present invention calculates after edge repair, and obtains a more accurate visual monitoring result of the traffic engineering construction equipment.

[0152] The embodiment of the present invention further provides a traffic engineering construction equipment safety visual monitoring device, the device comprising:

[0153] The first determination module is used to obtain an edge image of the traffic engineering construction equipment to be monitored and determine each to-be-determined edge pixel point in the edge image.

[0154] The second determination module is used to analyze the local deformation of the edge according to the position of each pending edge pixel point, determine the false deformation degree of each pending edge pixel point, and select each false deformed pixel point based on the false deformation degree; determine a number of repair reference pixel points corresponding to each false deformed pixel point, and the repair reference pixel points refer to a first preset number of pending edge pixel points located around the false deformed pixel point and on the same edge as the false deformed pixel point.

[0155] The third determination module is used to determine the restoration weight of each restoration reference pixel corresponding to each false deformed pixel according to the position of each false deformed pixel and each corresponding restoration reference pixel, the false deformation degree and gray value of each restoration reference pixel.

[0156] The fourth determination module is used to determine the restoration position of each false deformed pixel point according to the restoration weight and position of each restoration pixel point corresponding to each false deformed pixel point, so as to obtain a restored edge image.

[0157] The monitoring module is used to perform deformation monitoring on the traffic engineering construction equipment to be monitored based on the repaired edge image to obtain a deformation monitoring result.

[0158] In some possible implementations, the first determination module is also used to count the number of edge pixels adjacent to any edge pixel in the edge image in the eight-neighborhood direction of the edge pixel; if the number of edge pixels is greater than a preset pixel number threshold, the edge pixel is discarded; if the number of edge pixels is not greater than the preset pixel number threshold, the inclination angle change value of the edge pixel is determined based on the position of the edge pixel and the edge pixels located around the edge pixel; if the inclination angle change value of the edge pixel is greater than the angle change threshold, the edge pixel is discarded, otherwise, the edge pixel is used as a pending edge pixel.

[0159] In some possible implementations, the first determination module is also used to select the same number of edge pixel points on both sides of the edge to which the edge pixel point belongs as neighborhood pixel points, and use the neighborhood pixel points on one side of the edge pixel point as first neighborhood pixel points, and use the neighborhood pixel points on the other side of the edge pixel point as second neighborhood pixel points; determine the slope of the line between the edge pixel point and each first neighborhood pixel point as the first slope, and determine the slope of the line between the edge pixel point and each second neighborhood pixel point as the second slope; determine the inclination angle change value of the edge pixel point based on the difference between all the first slopes and all the second slopes corresponding to the edge pixel point.

[0160] In some possible implementations, the first determination module is also used to convert the average value of all first slopes corresponding to the edge pixel point into an angle value, recorded as the first angle value; convert the average value of all second slopes corresponding to the edge pixel point into an angle value, recorded as the second angle value; calculate the difference between the first angle value and the second angle value, and determine it as the inclination angle change value of the edge pixel point.

[0161] In some possible implementations, the second determination module is further used to take any undetermined edge pixel point as a target point, and select a second preset number of undetermined edge pixel points that are closest to the target point on the edge to which the target point belongs as adjacent pixel points of the target point; and use the variance of the slope of the line between the target point and each of its adjacent pixel points to analyze the significant change in the slope of the edge and determine the degree of false deformation of the target point.

[0162] In some possible implementations, the second determination module is further configured to set a false deformation threshold, and for all pending edge pixel points, the pending pixel points whose false deformation degree is greater than the false deformation threshold are regarded as false deformation pixel points.

[0163] In some possible implementations, the third determination module is further used to determine, for any false deformed pixel point, the distance between the false deformed pixel point and its corresponding repair reference pixels according to the positions of the false deformed pixel point and its corresponding repair reference pixels, and perform negative correlation normalization processing on the distance as the first repair weight factor of the corresponding repair reference pixel point; determine the repair weight of each repair reference pixel point corresponding to the false deformed pixel point according to the first repair weight factor, false deformation degree and grayscale value of each repair reference pixel point corresponding to the false deformed pixel point.

[0164] In some possible implementations, the third determination module is also used to perform negative correlation normalization processing on the degree of false deformation of any repair reference pixel point to obtain a second repair weight factor; calculate the difference between the grayscale value of the repair reference pixel point and the mode of the grayscale value on the edge to which the false deformed pixel point belongs, and perform negative correlation normalization processing on the difference value to obtain a third repair weight factor; fuse the first repair weight factor, the second repair weight factor and the third repair weight factor of the repair reference pixel point to obtain a fused value; calculate the proportion of the fused value in the cumulative value of the fused values ​​of all repair reference pixels corresponding to the false deformed pixel point, and determine the repair weight of the repair reference pixel point.

[0165] In some possible implementations, the fourth determination module is further used to perform weighted summation processing on any false deformed pixel point according to the restoration weight and position of each restoration pixel point corresponding to the false deformed pixel point, so as to obtain the restoration position of the false deformed pixel point.

[0166] In some possible implementations, the monitoring module is also used to perform curve fitting based on each edge pixel point on each edge in the repaired edge image to obtain each edge fitting curve; and determine whether the traffic engineering construction equipment to be monitored is deformed based on the curvature of each data point on each edge fitting curve.

[0167] Optionally, the transmission medium can be a wired link (such as but not limited to coaxial cable, optical fiber and digital subscriber line (DSL)) or a wireless link (such as but not limited to wireless Fidelity (WIFI), Bluetooth and mobile device network).

[0168] It should be noted that the device provided in the above embodiment is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device can be divided into different functional modules to complete all or part of the functions described above.

[0169] Figure 7 is a schematic diagram of the structure of a computer device provided by an embodiment of the present invention. Figure 7 As shown, the computer device 700 includes: a memory 701, a processor 702, and a computer program 703 stored in the memory 701 and running on the processor 702, wherein when the processor 702 executes the computer program 703, the computer device can execute any one of the aforementioned methods for visual safety monitoring of traffic engineering construction equipment.

[0170] In addition, an embodiment of the present invention also protects a device, which may include a memory and a processor, wherein the memory stores executable program code, and the processor is used to call and execute the executable program code to execute a method for safe visual monitoring of traffic engineering construction equipment provided in an embodiment of the present application.

[0171] In this embodiment, the functional modules of the device can be divided according to the above method example. For example, each functional module can be corresponded, or two or more functions can be integrated into one processing module. The above integrated module can be implemented in the form of hardware. It should be noted that the division of modules in this embodiment is schematic and is only a logical function division. There may be other division methods in actual implementation.

[0172] In the case of dividing each module according to each function, the device may also include a signal uploading module, a determining module, an adjusting module, etc. It should be noted that all relevant contents of each step involved in the above method embodiment can be referred to the functional description of the corresponding functional module, which will not be repeated here.

[0173] It should be understood that the device provided in this embodiment is used to execute the above-mentioned method for visual safety monitoring of traffic engineering construction equipment, and thus can achieve the same effect as the above-mentioned implementation method.

[0174] In the case of an integrated unit, the device may include a processing module and a storage module. When the device is applied to a device, the processing module may be used to control and manage the actions of the device. The storage module may be used to support the device to execute mutual program codes, etc. The processing module may be a processor or a controller, which may implement or execute various exemplary logic blocks, modules and circuits described in conjunction with the disclosure of the present application. The processor may also be a combination that implements a computing function, such as a combination of one or more microprocessors, a combination of digital signal processing (DSP) and a microprocessor, etc. The storage module may be a memory.

[0175] In addition, the device provided in the embodiments of the present application may specifically be a chip, component or module, and the chip may include a connected processor and memory; wherein the memory is used to store instructions, and when the processor calls and executes the instructions, the chip can execute a method for safe visual monitoring of traffic engineering construction equipment provided in the above embodiments.

[0176] This embodiment also provides a computer-readable storage medium, in which a computer program code is stored. When the computer program code is executed on a computer, the computer executes the above-mentioned related method steps to implement a method for visual safety monitoring of traffic engineering construction equipment provided in the above embodiment.

[0177] The present embodiment also provides a computer program product, when the computer program product is run on a computer, the computer executes the above-mentioned related steps to implement a traffic engineering construction equipment safety visual monitoring method provided by the above embodiment. Among them, the device, computer-readable storage medium, computer program product or chip provided in the present embodiment are all used to execute the corresponding method provided above, so the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding method provided above, and will not be repeated here. Through the description of the above implementation mode, the technicians in the relevant field can understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In practical applications, the above-mentioned function allocation can be completed by different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In the embodiments provided in the present application, it should be understood that the disclosed device and method can be implemented in other ways. For example, the device embodiment described above is only schematic, for example, the division of modules or units is only a logical function division, and there can be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, 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 may be an indirect coupling or communication connection through some interface, device or unit, which may be electrical, mechanical or other forms.

[0178] 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 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. A method for visually monitoring the safety of traffic engineering construction equipment, characterized in that: The following steps are involved: Acquire an edge image of the traffic engineering construction equipment to be monitored, and determine each to-be-determined edge pixel point in the edge image; Analyze the local deformation of the edge according to the position of each undetermined edge pixel point, determine the false deformation degree of each undetermined edge pixel point, and select each false deformation pixel point based on the false deformation degree; Determine a number of repair reference pixel points corresponding to each false deformed pixel point, wherein the repair reference pixel points refer to a first preset number of to-be-determined edge pixel points located around the false deformed pixel point and on the same edge as the false deformed pixel point; Determine the restoration weight of each restoration reference pixel corresponding to each false deformed pixel according to the position of each false deformed pixel and each restoration reference pixel corresponding thereto, the false deformation degree and the gray value of each restoration reference pixel; According to the restoration weight and position of each restoration pixel point corresponding to each false deformation pixel point, the restoration position of each false deformation pixel point is determined, thereby obtaining a restoration edge image; Based on the repaired edge image, deformation monitoring is performed on the traffic engineering construction equipment to be monitored to obtain a deformation monitoring result; The step of determining each undetermined edge pixel point in the edge image comprises: For any edge pixel point in the edge image, count the number of edge pixel points adjacent to the edge pixel point in the eight-neighborhood direction of the edge pixel point; If the number of edge pixels is greater than a preset pixel number threshold, the edge pixel is discarded; if the number of edge pixels is not greater than the preset pixel number threshold, the tilt angle change value of the edge pixel is determined according to the position of the edge pixel and the edge pixels located around the edge pixel; If the tilt angle change value of the edge pixel point is greater than the angle change threshold, the edge pixel point is discarded, otherwise, the edge pixel point is used as a pending edge pixel point; Analyzing the local deformation of the edge according to the position of each pending edge pixel point to determine the false deformation degree of each pending edge pixel point includes: Taking any undetermined edge pixel point as the target point, selecting a second preset number of undetermined edge pixel points closest to the target point on the edge to which the target point belongs as adjacent pixel points of the target point; According to the slope of the line between the target point and each of its adjacent pixels, the variance analysis is used to analyze the significant changes in the slope of the edge and determine the degree of false deformation of the target point.

2. A traffic engineering construction equipment safety visual monitoring method according to claim 1, characterized in that: The step of determining the tilt angle change value of the edge pixel point according to the positions of the edge pixel point and the edge pixel points located around the edge pixel point comprises: Selecting the same number of edge pixels on both sides of the edge to which the edge pixel belongs as neighborhood pixels, taking the neighborhood pixels on one side of the edge pixel as first neighborhood pixels, and taking the neighborhood pixels on the other side of the edge pixel as second neighborhood pixels; Determine the slope of the line between the edge pixel point and each first neighborhood pixel point as the first slope, and determine the slope of the line between the edge pixel point and each second neighborhood pixel point as the second slope; The tilt angle change value of the edge pixel point is determined according to the difference between all the first slopes and all the second slopes corresponding to the edge pixel point.

3. A traffic engineering construction equipment safety visual monitoring method according to claim 2, characterized in that: The step of determining the tilt angle change value of the edge pixel point according to the difference between all the first slopes and all the second slopes corresponding to the edge pixel point includes: Converting the average value of all first slopes corresponding to the edge pixel point into an angle value, which is recorded as a first angle value; Converting the average value of all second slopes corresponding to the edge pixel point into an angle value, which is recorded as a second angle value; The difference between the first angle value and the second angle value is calculated and determined as the tilt angle change value of the edge pixel point.

4. A traffic engineering construction equipment safety visual monitoring method according to claim 1, characterized in that: The step of selecting each false deformation pixel point based on the false deformation degree includes: A false deformation threshold is set, and for all pending edge pixels, the pending pixels whose false deformation degree is greater than the false deformation threshold are regarded as false deformation pixels.

5. A traffic engineering construction equipment safety visual monitoring method according to claim 1, characterized in that: The method of determining the restoration weight of each restoration reference pixel corresponding to each false deformed pixel according to the position of each false deformed pixel and each restoration reference pixel corresponding thereto, the false deformation degree and the gray value of each restoration reference pixel, comprises: For any false deformed pixel point, according to the positions of the false deformed pixel point and the corresponding repair reference pixel points, the distance between the false deformed pixel point and the corresponding repair reference pixel points is determined, and a negatively correlated normalization process is performed on the distance, which is used as a first repair weight factor of the corresponding repair reference pixel point; The restoration weight of each restoration reference pixel point corresponding to the false deformed pixel point is determined according to the first restoration weight factor, the false deformation degree and the gray value of each restoration reference pixel point corresponding to the false deformed pixel point.

6. A traffic engineering construction equipment safety visual monitoring method according to claim 5, characterized in that: The step of determining the restoration weight of each restoration reference pixel corresponding to the false deformed pixel according to the first restoration weight factor, the false deformation degree and the gray value of each restoration reference pixel corresponding to the false deformed pixel comprises: For any restoration reference pixel point, negatively correlated normalization processing is performed on the false deformation degree of the restoration reference pixel point to obtain a second restoration weight factor; Calculating the difference between the gray value of the repair reference pixel and the mode of the gray value on the edge to which the false deformed pixel belongs, performing negative correlation normalization processing on the difference to obtain a third repair weight factor; Fusing the first restoration weight factor, the second restoration weight factor, and the third restoration weight factor of the restoration reference pixel to obtain a fused value; The proportion of the fused value in the accumulated value of the fused values ​​of all the repair reference pixels corresponding to the false deformed pixel is calculated to determine the repair weight of the repair reference pixel.

7. A traffic engineering construction equipment safety visual monitoring method according to claim 1, characterized in that: The step of determining the repair position of each false deformed pixel point according to the repair weight and position of each repaired pixel point corresponding to each false deformed pixel point comprises: For any false deformed pixel point, a weighted summation process is performed according to the restoration weight and position of each restoration pixel point corresponding to the false deformed pixel point to obtain the restoration position of the false deformed pixel point.

8. A traffic engineering construction equipment safety visual monitoring method according to claim 1, characterized in that: The step of performing deformation monitoring on the traffic engineering construction equipment to be monitored based on the repaired edge image to obtain a deformation monitoring result includes: Performing curve fitting according to each edge pixel point on each edge in the repaired edge image to obtain each edge fitting curve; According to the curvature of each data point on each edge fitting curve, it is determined whether the traffic engineering construction equipment to be monitored is deformed.

Images