A method and system for intelligently measuring the size of a box beam template

Abstract

The present invention relates to the field of intelligent measurement technology, and in particular to a method and system for intelligently measuring the size of a box beam template. The method comprises: obtaining the first measurement coordinates of a projection point of a measuring device on the box beam template, classifying a plurality of first measurement coordinates to obtain two first measurement classification sets; performing plane fitting on the first measurement coordinates corresponding to data points in the two first measurement classification sets to obtain two first fitting planes; using an optimal plane partitioning model to classify a plurality of first measurement coordinates to obtain two second classification sets, calculating the mean distance between the data points in the second classification set and the partition plane in the optimal plane partitioning model, and taking the sum of the mean distances of the two second classification sets as the thickness of the box beam template. The present invention improves the working efficiency of measurement and the accuracy of measurement results.

Description

Technical Field

[0001] The present invention relates to the field of intelligent measurement technology, and in particular to a method and system for intelligently measuring the size of a box girder template. Background Art

[0002] Box girder is a common bridge structure, which is used in highway and railway bridges due to its good bearing capacity and bending resistance. Box girder formwork usually refers to the formwork system used for concrete box girder construction. The formwork is usually made of steel or aluminum alloy materials to ensure strength and stability. The formwork is usually made of steel or aluminum alloy materials to ensure strength and stability. The design of the box girder formwork must conform to the geometric shape of the box girder to ensure that the required structural shape can be formed after the concrete is poured. Therefore, when preparing the formwork, it is necessary to measure the size of the box girder formwork to ensure that it has the corresponding strength and structural stability and realizes the corresponding function. The box girder formwork size measurement mainly includes the measurement of length, width and thickness.

[0003] When measuring the thickness of the box girder formwork, it is usually measured at the boundary of the box girder formwork. It is often difficult to directly measure the thickness of the middle part of the vector formwork. It is difficult to achieve comparison of the same position point by placing laser measuring instruments at different positions, resulting in low measurement efficiency and low accuracy of the measurement results. Summary of the invention

[0004] In order to solve the problem of low accuracy of the measurement results of the box girder plate thickness, the present invention provides a size intelligent measurement method and system for a box girder template.

[0005] In a first aspect, the present invention provides a method for intelligently measuring the size of a box beam template, which adopts the following technical solution:

[0006] Acquire the first measurement coordinates of the projection point of the measuring device on the box beam template, move the measuring device to obtain a plurality of first measurement coordinates, and classify the plurality of first measurement coordinates to obtain two first measurement classification sets;

[0007] Performing plane fitting on the first measurement coordinates corresponding to the measurement data points in the two first measurement classification sets to obtain two first fitting planes, and calculating the first difference between the two first measurement planes, where the first difference indicates the degree of difference between the two first measurement planes;

[0008] Using the optimal plane division model to classify multiple first measurement coordinates to obtain two second measurement classification sets, calculating the mean distance between the measurement data points in the second measurement classification set and the division plane in the optimal plane division model, and taking the sum of the distance means of the two second measurement classification sets as the measurement thickness of the box girder template;

[0009] Among them, the method for obtaining the optimal plane division model is: constructing an initial plane division model, using the initial plane division model to classify multiple first measurement coordinates to obtain two data sets, performing plane fitting on the first measurement coordinates in the two data sets to obtain two second fitting planes; calculating the second difference of the two second fitting planes; using the first difference and the second difference to construct an optimization function, and solving the optimization function to obtain the optimal plane division model.

[0010] The effect is that by obtaining the first measurement coordinates, the first measurement coordinates are classified and fitted using the optimal plane division model, and finally the thickness size of the box girder plate is obtained. Compared with the traditional measurement method, the measurement efficiency and the accuracy of the measurement results are improved.

[0011] Preferably, the expression of the first difference is:

[0012]

[0013] In the formula, Indicates the first difference, is the normal vector corresponding to one of the first fitting planes, is the normal vector corresponding to the other first fitting plane, for , The average vector of is the vector corresponding to the data mean point of the two planes, is an exponential function with base e.

[0014] Preferably, the expression of the first difference is:

[0015]

[0016] In the formula, Indicates the first difference, is the normal vector corresponding to one of the first fitting planes, is the normal vector corresponding to the other first fitting plane, for , The average vector of is the vector corresponding to the data mean point of the two planes.

[0017] Preferably, the expression of the first difference is:

[0018]

[0019] In the formula, Indicates the first difference, is the normal vector corresponding to one of the first fitting planes, is the normal vector corresponding to the other first fitting plane, for , The average vector of is the vector corresponding to the data mean point of the two planes, is an exponential function with base e.

[0020] The effect is that, by calculating the first difference, the fitting effect of the first fitting plane can be judged, thereby further improving the accuracy of the thickness measurement result.

[0021] Preferably, the method for the mobile measuring device to obtain a plurality of first measurement coordinates is:

[0022] The measuring device is located on one side of the box beam template, and is moved along the length direction of the box beam template to obtain multiple first measurement coordinates; the measuring device is located on the other side of the box beam template, and is repeatedly moved along the length direction of the box beam template to obtain multiple first measurement coordinates again.

[0023] The effect is that a plurality of first measurement coordinates can be quickly obtained by moving the measuring device, thereby improving the work efficiency of obtaining the first measurement coordinates and saving the time required for measuring the plate thickness.

[0024] Preferably, the plane partitioning model is an SVM model, and the hyperplane in the SVM model is used as the partitioning plane.

[0025] Preferably, before plane fitting is performed on the first measurement coordinates corresponding to the data points in the two data sets, the method further includes the steps of detecting and deleting abnormal data points.

[0026] The effect is that by deleting abnormal data points, it is possible to avoid interference of abnormal data on the fitting plane, thereby improving the fitting effect.

[0027] Preferably, the expression of the optimization function is: ; In the formula, Indicates the first difference, Indicates the second difference.

[0028] In a second aspect, the present invention provides a box beam template size intelligent measurement system, which adopts the following technical solution:

[0029] A box girder template size intelligent measurement system, a processor and a memory, the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned box girder template size intelligent measurement method is implemented.

[0030] The beneficial effect is: the above-mentioned intelligent measurement method for the size of a box beam template is generated into a computer program and stored in a memory so as to be loaded and executed by a processor, thereby making a system based on the memory and the processor for easy use.

[0031] The present invention has the following technical effects:

[0032] By obtaining the first measurement coordinates, the SVM model is used to classify and fit the first measurement coordinates to obtain the second fitting plane, and the thickness of the box girder plate is obtained according to the distance between the data points in the second classification set and the hyperplane in the SVM model. Compared with the traditional measurement method, the time required for data collection is saved, and the measurement efficiency and the accuracy of the measurement results are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] By reading the detailed description below with reference to the accompanying drawings, the above and other purposes, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-restrictive manner, and the same or corresponding reference numerals represent the same or corresponding parts.

[0034] Figure 1 The present invention is a method flow chart of a method for intelligently measuring the dimensions of a box beam template.

[0035] Figure 2 It is a schematic diagram of the box beam template structure of the present invention.

[0036] In the figure: 10. Box girder formwork; 20. Measuring device. DETAILED DESCRIPTION

[0037] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0038] It should be understood that when the terms "first", "second", etc. are used in the claims, descriptions, and drawings of the present invention, they are only used to distinguish different objects, rather than to describe a specific order. The terms "include" and "comprise" used in the description and claims of the present invention indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their collections.

[0039] like Figure 2As shown, the object to be measured in the present invention is the thickness dimension of the box beam template 10. In the present invention, the direction perpendicular to the paper surface is the length direction of the box beam template 10, and the measuring device 20 is used to obtain the second coordinate of the measuring device 20 itself and the distance between the measuring device 20 and one side of the box beam template 10, and then the first measurement coordinate of the projection point of the measuring device 20 on the side of the box beam template 10 is calculated. The measuring device 20 adopts a handheld laser rangefinder, in which the projection point can also be understood as the irradiation point of the laser of the measuring device 20, and the following embodiments are described based on this background.

[0040] The embodiment of the present invention discloses a method for intelligently measuring the size of a box beam template, referring to Figure 1 , including the following steps, as follows:

[0041] S1: Obtain the first measurement coordinates of the projection point of the measuring device on the box beam template, and move the measuring device to obtain multiple first measurement coordinates.

[0042] The spatial coordinates of the measuring device are obtained, and the spatial coordinates of the measuring device are used as the second coordinates. The measuring device is located on one side of the box beam template, and the measuring device is moved along the length direction of the box beam template to obtain multiple first measurement coordinates and multiple second coordinates; the measuring device is located on the other side of the box beam template, and the measuring device is repeatedly moved along the length direction of the box beam template to obtain multiple first measurement coordinates and multiple second coordinates again. The first measurement coordinate and the second measurement coordinate are both three-dimensional spatial coordinates, and each first measurement coordinate corresponds to a second coordinate. In this embodiment, the measuring device is a handheld laser rangefinder, and the distance between the handheld laser rangefinder body and one side of the box beam template is measured by the handheld laser rangefinder. Then, according to the three-dimensional spatial coordinates (second coordinates) of the handheld laser rangefinder, the three-dimensional spatial coordinates (first measurement coordinates) of the projection point of the handheld laser rangefinder on the box beam template can be obtained, and the projection point of the handheld laser rangefinder on the box beam template is used as the measurement point.

[0043] S2: Classify the multiple first measurement coordinates to obtain two first measurement classification sets.

[0044] The multiple first measurement coordinates obtained in step S1 are clustered using the DBSCAN algorithm to obtain multiple cluster clusters, and the first measurement coordinates are classified to facilitate subsequent analysis of the first measurement coordinates of the same category. In this embodiment, the minimum number of samples during classification is 3, and the cluster radius is 0.05. The minimum number of samples and the cluster radius can also be adjusted according to actual conditions.

[0045] The second coordinates corresponding to the data points in any of the clusters are classified into two subsets using the k-means algorithm; the second coordinates are the three-dimensional spatial coordinates of the handheld laser rangefinder. Among the obtained second coordinates, part of the second coordinates are the coordinates when the handheld laser rangefinder is located on one side of the box beam template, and the other part of the second coordinates are the coordinates when the handheld laser rangefinder is located on the other side of the box beam template. Therefore, the second coordinates are clustered using the k-means algorithm to obtain two subsets, one of which corresponds to one side. It can be understood that a second coordinate in a subset corresponds to a first measurement coordinate in the corresponding cluster. Therefore, the data in the corresponding cluster is divided into two first measurement classification sets according to the two subsets.

[0046] Exemplarily, the first measured coordinate point in the cluster is ( , , , , , ), the second coordinate point corresponding to the cluster is ( , , , , , ), the k-means algorithm is used to cluster the second coordinate to obtain two subsets ( , , )and , ), the clusters are divided into two first measurement classification sets according to the two subsets ( , , )and( , ),in( , ) is the data point on one side of the box girder formwork, ( , , ) is the data point on the other side of the box beam template. It should be noted that, for ease of understanding, it is assumed that there are 4 cluster data points. In practical applications, there is no limit on the number of data points in the cluster.

[0047] S3: performing plane fitting on the first measurement coordinates corresponding to the data points in the two first measurement classification sets to obtain two first fitting planes, and calculating a first difference between the two first fitting planes, where the first difference indicates a degree of difference between the two first fitting planes.

[0048] Exemplarily, for the first measurement classification set obtained in step S2 ( , , ), a first fitting plane is obtained by fitting using the least squares plane fitting algorithm, which can also be understood as a side of the box girder template. Similarly, for the first measurement classification set ( , ) is plane fitted to obtain another first fitting plane, which can also be understood as the other side of the box girder formwork. The distance between the two first fitting planes is the thickness of the box girder plate.

[0049] S4: Calculate a first difference between the two first fitting planes, where the first difference indicates a degree of difference between the two first fitting planes.

[0050] In one embodiment, the expression of the first difference is:

[0051]

[0052] In the formula, For the first difference, is the normal vector corresponding to one of the first fitting planes, is the normal vector corresponding to the other first fitting plane, for , The average vector of is the vector corresponding to the data mean point of the two planes, is an exponential function with base e, For vector exist Projection value onto the normal vector.

[0053] In the formula, Representation vector , The cosine of the angle between them. When the cosine value is 1, it means that the vector and coincide with each other and have the same direction; a cosine value of -1 indicates that the vector and coincident, opposite in direction; a cosine value of 0 indicates that the vector and Independent of each other. The value of can determine whether the first fitting plane is parallel to the second fitting plane. It can be understood that when The closer it is to 1, the better the parallelism between the first fitting plane and the second fitting plane is, and the better the fitting effect between the first fitting plane and the second fitting plane is.

[0054] for , The average vector of represents the direction of the angle between the two first fitting planes; is the vector corresponding to the data mean point of the two first fitting planes. The mean point of the first fitting plane is: The axis average value is obtained , for the data points on the first fitting plane The axis average value is obtained , for the data points on the first fitting plane The axis average value is obtained , the coordinates of the data mean point are ( , , ), the two corresponding data mean points of the two first fitting planes, connecting the two data mean points is ,if and Vertical, that is If the value of is 0, it means that the two first fitting planes are parallel. If they are not perpendicular, it means that there is a certain angle between the two first fitting planes. For vector exist The projection value of the normal vector, specifically the vector The modulus and angle ( If the projection value is 0, it means that the two faces coincide.

[0055] In summary, we can know The closer the value of is to 1, the better the overlap between the two first fitting planes is, which further indicates that the fitting effect of the first fitting plane is better. If it is closer to 0, it means that the overlap between the two first fitting planes is worse, which further indicates that the fitting effect of the first fitting plane is worse.

[0056] In another embodiment, the expression of the first difference is:

[0057]

[0058] In the formula, is the normal vector corresponding to one of the first fitting planes, is the normal vector corresponding to the other first fitting plane, for , The average vector of is the vector corresponding to the data mean point of the two planes.

[0059] In yet another embodiment, the expression of the first difference is:

[0060]

[0061] In the formula, is the normal vector corresponding to one of the first fitting planes, is the normal vector corresponding to the other first fitting plane, for , The average vector of is the vector corresponding to the data mean point of the two planes, is an exponential function with base e.

[0062] S5: construct an initial plane division model, use the initial plane division model to classify multiple first measurement coordinates to obtain two data sets, perform plane fitting on the data points in the two data sets to obtain two second fitting planes, and calculate the second difference between the two second fitting planes, where the second difference represents the degree of difference between the two second fitting planes.

[0063] The plane partition model is an SVM model, specifically: constructing an initial SVM model, the dimensionality-increasing kernel function of the initial SVM model adopts a Gaussian kernel function, and using the initial SVM model to classify the second coordinates corresponding to the data points in the cluster cluster to obtain two classification sets. It should be noted that the cluster cluster here is the same cluster cluster as the cluster cluster for binary classification using the k-means algorithm in step S2. The parameters of the dimensionality-increasing kernel function are different, and the dimensionality-increasing effect and classification effect are also different. Therefore, by adjusting the parameters of the dimensionality-increasing kernel function, the dimensionality-increasing effect of the SVM model is adjusted, so that the classification result of the SVM model should be the same as the classification result of the k-means algorithm. Then, outlier detection is performed on the data in the classification set to remove outliers, and the classification set is fitted separately using the least squares plane fitting algorithm to obtain two second fitting planes. The Euclidean distance from the data in the classification set to the corresponding second fitting plane is calculated, and then the Euclidean distance obtained is tested for outliers using the triple standard deviation method. The abnormal data in the classification set is then deleted, and the remaining data is plane fitted again to obtain the second fitting plane. The second difference between the two second fitting planes is calculated. The calculation method of the second difference is the same as the calculation method of the first difference, which will not be repeated here.

[0064] S6: construct an optimization function using the first difference and the second difference, and solve the optimization function to obtain an optimal plane partitioning model.

[0065] The expression of the optimization function is: ; In the formula, Indicates the first difference, Indicates the second difference; the objective function is maximized by the group optimization intelligent algorithm to obtain the adaptive optimal kernel function, the SVM model corresponding to the adaptive optimal kernel function is used as the optimal plane partitioning model, and the hyperplane in the SVM model corresponding to the adaptive optimal kernel function is used as the partitioning plane. The maximum value is solved to improve The lower limit of .

[0066] S7: Use the optimal plane division model to classify multiple first measurement coordinates to obtain two second measurement classification sets, calculate the mean distance between the measurement data points in the second measurement classification set and the division plane in the optimal plane division model, and use the sum of the distance means of the two second measurement classification sets as the measured thickness of the box girder formwork.

[0067] The optimal plane partitioning model is used to classify the data points in the cluster to obtain two second classification sets. The cluster here is the same cluster as the cluster that is classified by the k-means algorithm in step S2. The hyperplane of the SVM model is the middle reference plane of the box girder formwork, which can also be understood as the middle virtual reference plane of 1 / 2 plate thickness. The distance between each data point in the second classification set and the hyperplane in the SVM model is calculated, and finally the distance mean of the data points in the second classification set is obtained. The two second classification sets correspond to two distance means, and the sum of the two means is the thickness of the box girder formwork.

[0068] An embodiment of the present invention further discloses a box girder template size intelligent measurement system, including a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, an intelligent box girder template size measurement method according to the present invention is implemented.

[0069] The above system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface, and their configuration and functions are known in the art, so they will not be described in detail here.

[0070] In the present invention, the aforementioned memory may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, apparatus or device. For example, a computer-readable storage medium may be any appropriate magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory RRAM (Resistive Random Access Memory), a dynamic random access memory DRAM (Dynamic Random Access Memory), a static random access memory SRAM (Static Random Access Memory), an enhanced dynamic random access memory EDRAM (Enhanced Dynamic Random Access Memory), a high bandwidth memory HBM (High Bandwidth Memory), a hybrid memory cube HMC (Hybrid Memory Cube), etc., or any other medium that can be used to store the required information and can be accessed by an application, a module or both. Any such computer storage medium may be part of a device or accessible or connectable to a device.

[0071] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

[0072] The above are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for intelligently measuring the size of a box beam template, characterized in that: Includes steps: The first measurement coordinate of the projection point of the measuring device on the box beam template is obtained, the measuring device is moved to obtain multiple first measurement coordinates, and the multiple first measurement coordinates are classified to obtain two first measurement classification sets; the method for classifying the multiple first measurement coordinates to obtain the two first measurement classification sets is: the second coordinates corresponding to each first measurement coordinate are classified into two categories by using the k-means algorithm, so that the second coordinates corresponding to each first measurement coordinate are divided into two categories, and the first measurement coordinates corresponding to the second coordinates in the same category are classified into the same classification set; the second coordinate corresponding to the first measurement coordinate refers to the three-dimensional space coordinate of the handheld laser rangefinder when the first measurement coordinate is collected; Performing plane fitting on the first measurement coordinates corresponding to the measurement data points in the two first measurement classification sets to obtain two first fitting planes, and calculating the first difference between the two first measurement planes, where the first difference indicates the degree of difference between the two first measurement planes; The optimal plane partitioning model is used to classify multiple first measurement coordinates to obtain two second measurement classification sets, the distance mean between the measurement data points in the second measurement classification set and the division plane in the optimal plane partitioning model is calculated, and the sum of the distance means of the two second measurement classification sets is used as the measurement thickness of the box girder template; the plane partitioning model is an SVM model, and the hyperplane in the SVM model is used as the division plane; The method for obtaining the optimal plane partition model is as follows: constructing an initial plane partition model, using the initial plane partition model to classify multiple first measurement coordinates to obtain two data sets, performing plane fitting on the first measurement coordinates in the two data sets to obtain two second fitting planes; calculating the second difference of the two second fitting planes; constructing an optimization function using the first difference and the second difference, solving the optimization function to obtain the optimal plane partition model; solving the optimization function includes: solving the optimization function for the maximum value through a group optimization intelligent algorithm to obtain an adaptive optimal kernel function, and using the SVM model corresponding to the adaptive optimal kernel function as the optimal plane partition model; The expression of the first difference is: In the formula, Indicates the first difference, is the normal vector corresponding to one of the first fitting planes, is the normal vector corresponding to the other first fitting plane, for , The average vector of is the vector corresponding to the data mean point of the two planes, is an exponential function with base e; The expression of the optimization function is: ; In the formula, Indicates the first difference, Indicates the second difference; Or the expression of the first difference is: In the formula, Indicates the first difference, is the normal vector corresponding to one of the first fitting planes, is the normal vector corresponding to the other first fitting plane, for , The average vector of is the vector corresponding to the data mean point of the two planes; Or the expression of the first difference is: In the formula, Indicates the first difference, is the normal vector corresponding to one of the first fitting planes, is the normal vector corresponding to the other first fitting plane, for , The average vector of is the vector corresponding to the data mean point of the two planes, is an exponential function with base e; The method for obtaining a plurality of first measurement coordinates by the mobile measurement device is: The measuring device is located on one side of the box beam template, and is moved along the length direction of the box beam template to obtain multiple first measurement coordinates; the measuring device is located on the other side of the box beam template, and is repeatedly moved along the length direction of the box beam template to obtain multiple first measurement coordinates again.

2. The intelligent measurement method for the size of a box beam template according to claim 1 is characterized in that: Before plane fitting is performed on the first measurement coordinates corresponding to the data points in the two data sets, the steps of detecting and deleting abnormal data points are also included.

3. An intelligent measurement system for the size of a box beam template, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a method for intelligently measuring the size of a box girder template according to claim 1 or 2 is implemented.

Images