CN118049938A

CN118049938A - Rail measuring method, device, electronic equipment, readable storage medium and system

Info

Publication number: CN118049938A
Application number: CN202410445698.1A
Authority: CN
Inventors: 王玮; 张睿; 庞天吉; 胡峻毅; 刘闯; 何贤昆
Original assignee: Xi'an Glasssix Network Technology Co ltd
Current assignee: Xi'an Glasssix Network Technology Co ltd
Priority date: 2024-04-15
Filing date: 2024-04-15
Publication date: 2024-05-17
Anticipated expiration: 2044-04-15
Also published as: CN118049938B

Abstract

The invention relates to the technical field of steel rail measurement, and provides a steel rail measurement method, a device, electronic equipment, a readable storage medium and a system, wherein the method comprises the following steps: acquiring a profile data segment and an item to be measured; selecting a plurality of segments of target profile data; extracting geometric features from the multiple segments of target contour data segments, the geometric features comprising a first geometric feature of each target contour data segment or comprising a first geometric feature of each target contour data segment and at least one second geometric feature, each first geometric feature being determined by the geometry of the corresponding target contour data segment, each second geometric feature being determined by the geometric relationship of two reference geometric features, each reference geometric feature being a first geometric feature or another second geometric feature that has been determined and that is different from the second geometric feature that needs to be determined; selecting a target geometric feature from the geometric features; and calculating measurement data according to the target geometric features and the measurement strategy. The invention can improve the measurement efficiency.

Description

Rail measuring method, device, electronic equipment, readable storage medium and system

Technical Field

The invention relates to the technical field of steel rail measurement, in particular to a steel rail measurement method, a device, electronic equipment, a readable storage medium and a system.

Background

The contour shape of the steel rail is an important geometric parameter of the steel rail, and the contour shape error has an important influence on the product quality of the steel rail. With the development of optical and computer technologies in recent years, a non-contact measurement scheme based on visual and line laser has been widely applied to a rail profile measurement link. According to different measurement requirements, the vision scheme is mainly used for detecting the surface defects of the steel rail, and the line laser scheme is mainly used for high-precision profile shape measurement.

In a high-precision steel rail profile shape measurement task, a plurality of linear laser sensors are generally adopted as data acquisition units, the sensors are calibrated to obtain installation parameters of each sensor, a transformation matrix is constructed by using the installation parameters, and the measurement coordinate system of each laser sensor is aligned through coordinate transformation to obtain complete steel rail section profile data after data fusion.

In the prior art, after the profile data of the steel rail is obtained, the measurement data of the steel rail is generally directly calculated according to the geometric features of the profile data, but for some measurement requirements, the measurement data cannot meet the measurement requirements, and in this case, the measurement data is also required to be processed again, so that the measurement efficiency is affected.

Disclosure of Invention

The invention aims to provide a steel rail measuring method, a device, electronic equipment, a readable storage medium and a system, which can improve the measuring efficiency.

Embodiments of the invention may be implemented as follows:

In a first aspect, the present invention provides a rail measurement method, the method comprising:

acquiring a profile data segment of a steel rail to be measured and a project to be measured, wherein the profile data segment is obtained by dividing a profile point cloud of the steel rail to be measured, and the project to be measured corresponds to a measurement strategy;

selecting a plurality of segments of target contour data segments from the contour data segments according to the item to be measured;

Extracting geometric features from the multiple segments of target contour data segments, wherein the geometric features comprise first geometric features of each target contour data segment or comprise first geometric features and at least one second geometric feature of each target contour data segment, each first geometric feature is determined by the geometric form of the corresponding target contour data segment, each second geometric feature is determined by the geometric relation of two reference geometric features, and each reference geometric feature is a first geometric feature or other determined and different from the second geometric feature required to be determined;

Selecting target geometric features meeting the visual measurement requirements of the item to be measured from the geometric features;

And calculating the measurement data of the item to be measured according to the target geometric features and the measurement strategy.

In an alternative embodiment, each of the target profile data segments corresponds to a first extraction strategy that meets the visual measurement requirement of the item to be measured, and the step of extracting geometric features from the multiple segments of target profile data segments includes:

For any contour data segment to be extracted in the target contour data segment, if the first extraction strategy is feature point extraction, taking feature points of contour point clouds in the contour data segment to be extracted as the first geometric features;

and if the first extraction strategy is geometric fitting, fitting all contour point clouds of the contour data segment to be extracted according to the geometric form of the contour data segment to be extracted to obtain the first geometric feature.

In an optional embodiment, the step of fitting all contour point clouds of the contour data segment to be extracted according to the geometry of the contour data segment to be extracted to obtain the first geometric feature includes:

Determining an error equation of the profile data segment to be extracted according to the geometric form of the profile data segment to be extracted;

determining geometric parameters of the error equation according to all contour point clouds of the contour data segment to be extracted;

And taking the error equation after the geometric parameter is determined as the first geometric feature.

In an alternative embodiment, the step of extracting geometric features from the multi-segment target contour data segment further comprises:

Obtaining an extraction configuration meeting the visual measurement requirement of the item to be measured, wherein the extraction configuration comprises a second extraction strategy, a first reference geometric feature and a second reference geometric feature;

A second geometric feature is extracted from the first reference geometric feature and the second reference geometric feature.

In an alternative embodiment, the step of extracting a second geometric feature from the first reference geometric feature and the second reference geometric feature comprises:

if the second extraction strategy is intersection point extraction, calculating intersection points of the first reference geometric features and the second reference geometric features, and taking the calculated intersection points as the second geometric features;

If the second extraction strategy is linear extraction, calculating a characteristic straight line determined by the first reference geometric feature and the second reference geometric feature, and taking the characteristic straight line as the second geometric feature.

In an alternative embodiment, the step of calculating a feature straight line determined by the first reference geometrical feature and the second reference geometrical feature comprises:

If the first reference geometric feature is a point, the second reference geometric feature is a straight line, and the first reference geometric feature is not located on the second reference geometric feature, determining a straight line passing through the first reference geometric feature and parallel to the second reference geometric feature as the feature straight line;

If the first reference geometric feature and the second reference geometric feature belong to a symmetrical geometric relationship, determining a symmetrical center line of the first reference geometric feature and the second reference geometric feature as the feature straight line;

And if the first reference geometric feature is a point and the second reference geometric feature is a point, determining a straight line of the first reference geometric feature and the second reference geometric feature as the feature straight line.

In an alternative embodiment, the target geometric features are two, and the step of calculating the measurement data of the item to be measured according to the target geometric features and the measurement strategy includes:

If the measurement strategy is distance measurement, calculating the distance between the two target geometric features, and taking the calculated distance as measurement data;

if the measurement strategy is angle measurement, calculating an included angle between the two target geometric features, and taking the calculated angle as measurement data.

In a second aspect, the present invention provides a rail measurement device, the device comprising:

The acquisition module is used for acquiring a profile data segment of the steel rail to be measured and a project to be measured, wherein the profile data segment is obtained by dividing a profile point cloud of the steel rail to be measured, and the project to be measured corresponds to a measurement strategy;

a selecting module, configured to select a plurality of segments of target contour data segments from the contour data segments according to the item to be measured;

An extraction module, configured to extract geometric features from the multiple segments of target contour data segments, where the geometric features include a first geometric feature of each of the target contour data segments or include a first geometric feature of each of the target contour data segments and at least one second geometric feature, where each of the first geometric features is determined by a geometric form of a corresponding target contour data segment, and each of the second geometric features is determined by a geometric relationship between two reference geometric features, and each of the reference geometric features is a first geometric feature or another determined and different second geometric feature than the second geometric feature that needs to be determined;

the selection module is further used for selecting target geometric features meeting the visual measurement requirements of the item to be measured from the geometric features;

And the calculation module is used for calculating the measurement data of the item to be measured according to the target geometric characteristics and the measurement strategy.

In a third aspect, the present invention provides an electronic device comprising a processor and a memory, the memory being for storing a program, the processor being for implementing the rail measurement method of any one of the preceding embodiments when the program is executed.

In a fourth aspect, the present invention provides a computer readable storage medium having stored thereon a computer program which when executed by a processor implements a rail measurement method as in any of the preceding embodiments.

In a fifth aspect, the present invention provides a rail measurement system, where the rail measurement system includes a plurality of nodes organized based on a graph model according to an item to be measured, the plurality of nodes include data nodes, feature nodes, and measurement nodes, each of the nodes is a vertex of the graph model, and an input-output relationship between two nodes is an edge between corresponding vertices of the graph model;

The data node is used for acquiring a profile data segment and a to-be-measured item of a steel rail to be measured, selecting a plurality of segments of target profile data segments from the profile data segment according to the to-be-measured item, inputting the plurality of segments of target profile data segments into the characteristic node, wherein the profile data segments are obtained by dividing a profile point cloud of the steel rail to be measured, and the to-be-measured item corresponds to a measurement strategy;

The feature node is configured to extract geometric features from the multiple segments of target contour data segments and select target geometric features from the geometric features that meet the visual measurement requirement of the item to be measured, and input the target geometric features to the measurement node, where the geometric features include a first geometric feature of each of the target contour data segments or include a first geometric feature of each of the target contour data segments and at least one second geometric feature, each of the first geometric features is determined by a geometric shape of a corresponding target contour data segment, each of the second geometric features is determined by a geometric relationship of two reference geometric features, and each of the reference geometric features is a first geometric feature or a determined second geometric feature different from the second geometric feature that needs to be determined;

And the measurement node is used for calculating the measurement data of the item to be measured according to the target geometric feature and the measurement strategy.

Compared with the prior art, after a plurality of segments of target contour data segments are selected from the contour data segments according to the item to be measured, the first geometric feature of each target contour data segment is extracted from the plurality of segments of target contour data segments, or the first geometric feature of each target contour data segment is extracted and at least one second geometric feature is extracted, each first geometric feature is determined by the geometric form of the corresponding target contour data segment, each second geometric feature is determined by the geometric relation of two reference geometric features, and each reference geometric feature is the first geometric feature or other determined second geometric feature which is different from the second geometric feature to be determined; selecting target geometric features meeting the visual measurement requirements of the item to be measured from the geometric features; and finally, calculating the measurement data of the item to be measured according to the target geometric characteristics and the measurement strategy. According to the invention, the first geometric feature or the first geometric feature and the second geometric feature which meet the measurement requirement are automatically extracted, and the first geometric feature characterizes the geometric form of the target outline data segment and the second geometric feature characterizes the geometric relation of the target outline data segment, so that the target geometric feature which meets the measurement requirement is selected from the first geometric feature or the first geometric feature and the second geometric feature, and further, the measurement data of the item to be measured is directly calculated according to the target geometric feature and the measurement strategy, and the situation that the measurement data obtained by direct calculation is required to be reprocessed when the measurement requirement cannot be met is avoided, thereby improving the measurement efficiency.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the embodiments will be briefly described below, it being understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and other related drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

Fig. 1 is a flowchart illustrating a rail measurement method according to the present embodiment.

Fig. 2 is an exemplary diagram of a profile tag of a profile point cloud of a rail to be tested according to the present embodiment.

Fig. 3 is an exemplary diagram of a rail height measurement process of the rail provided in the present embodiment.

Fig. 4 is an exemplary diagram of a rail head width measurement process of the steel rail provided in the present embodiment.

Fig. 5 is an exemplary diagram of a rail measurement overall architecture according to the present embodiment.

Fig. 6 is a block diagram of the rail measuring device according to the present embodiment.

Fig. 7 is a block diagram of an electronic device according to the present embodiment.

Icon: 10-an electronic device; 11-a processor; 12-memory; 13-bus; 100-a steel rail measuring device; 110-an acquisition module; 120-a selection module; 130-an extraction module; 140-calculation module.

Detailed Description

For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. The components of the embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations.

Thus, the following detailed description of the embodiments of the invention, as presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

It should be noted that: like reference numerals and letters denote like items in the following figures, and thus once an item is defined in one figure, no further definition or explanation thereof is necessary in the following figures.

In the description of the present invention, it should be noted that, if the terms "upper", "lower", "inner", "outer", and the like indicate an azimuth or a positional relationship based on the azimuth or the positional relationship shown in the drawings, or the azimuth or the positional relationship in which the inventive product is conventionally put in use, it is merely for convenience of describing the present invention and simplifying the description, and it is not indicated or implied that the apparatus or element referred to must have a specific azimuth, be configured and operated in a specific azimuth, and thus it should not be construed as limiting the present invention.

Furthermore, the terms "first," "second," and the like, if any, are used merely for distinguishing between descriptions and not for indicating or implying a relative importance.

It should be noted that the features of the embodiments of the present invention may be combined with each other without conflict.

Referring to fig. 1, fig. 1 is a flowchart illustrating a rail measurement method according to the present embodiment, and the method includes the following steps:

step S101, acquiring a profile data segment of a steel rail to be measured and a project to be measured, wherein the profile data segment is obtained by dividing a profile point cloud of the steel rail to be measured, and the project to be measured corresponds to a measurement strategy.

In this embodiment, a method for dividing the contour point cloud of the steel rail to be tested is as follows:

firstly, collecting contour point clouds of a rail to be tested and preprocessing;

The acquisition of the contour point cloud may be: and scanning the steel rail by using a line laser sensor to obtain the section data of each surface of the steel rail. And (3) calibrating the installation position of the line laser sensor, and converting the data under the coordinate system of each sensor into a unified coordinate system by using the installation parameters of the sensor, so as to realize contour splicing and obtain complete steel rail contour data. The preprocessing of the contour point cloud may be: and noise and abnormal values of the steel rail profile data are processed through filtering, interference of environment and hardware performance on measurement is reduced, and quality of acquired data is optimized.

Secondly, analyzing the steel rail CAD model to obtain standard template point cloud data;

When the CAD (Computer-AIDED DESIGN) model of the steel rail is analyzed, the analysis codes of the dxf file and the dxf file of the CAD model can be utilized to analyze the CAD model, the outline of the CAD model is analyzed into a series of straight line and circular arc parameters, point cloud data are generated according to parameter interpolation obtained by analysis, and the generated point cloud data are marked by utilizing different parameters to form standard template point cloud data with outline labels. Referring to fig. 2, fig. 2 is an exemplary diagram of an outline tag of standard template point cloud data according to the present embodiment. In fig. 2, the point cloud data of the same contour label belongs to the same straight line or the same arc, for example, the point cloud data of the contour label 17 belongs to the corresponding straight line, the point cloud data of the contour label 16 belongs to the corresponding arc, and other contour labels are similar.

Thirdly, registering the contour point cloud of the steel rail to be detected and the standard template point cloud data to obtain a contour label of the contour point cloud of the steel rail to be detected;

And registering the steel rail contour data and the standard template point cloud data by adopting an ICP (ITERATIVE CLOSEST POINT) algorithm, and marking the contour label of the steel rail contour point cloud closest to each standard template point cloud as the contour label of the same standard template after registering. The profile tag of the rail profile point cloud is similar to that of a standard template, and specific reference can be made to fig. 2.

And finally, organizing the contour point clouds of different contour labels in the same subset according to the requirements, wherein the contour point clouds in the same subset are one contour data segment, so as to realize the segmentation of the contour point clouds of the steel rail to be detected.

And placing the rail point clouds with different profile labels into the same subset to realize the segmentation of the profile data. For example, for fig. 2, the profile point clouds of the steel rail to be tested of the profile tag 4 and the profile tag 7 are divided into the same subset, corresponding to one profile data segment. These rail point cloud subsets with different profile labels are rail profile data segments, whereas the collection of rail point cloud subsets is the complete rail profile. The profile data segments to be selected are different from profile measurement items, and the required profile data segments can be selected from the profile point cloud of the steel rail to be measured marked with the profile label according to measurement requirements.

In this embodiment, different items to be measured correspond to different measurement strategies, for example, for the calculation of the rail height, the measurement strategy is the calculation distance, for the rail cross-sectional profile, the measurement strategy is the calculation angle, etc.

Step S102, selecting a plurality of segments of target contour data segments from the contour data segments according to the item to be measured.

In this embodiment, the profile data segments required by different items to be measured may be different, for example, when the items to be measured are track heights, the target profile data segments required to be selected may be the profile data segments composed of profile tags 4 and 7 and the profile data segments composed of profile tags 18 and 25 in fig. 2.

Step S103, extracting geometric features from the multiple segments of target contour data segments, wherein the geometric features include a first geometric feature of each target contour data segment or include a first geometric feature of each target contour data segment and at least one second geometric feature, each first geometric feature is determined by the geometric form of the corresponding target contour data segment, each second geometric feature is determined by the geometric relationship of two reference geometric features, and each reference geometric feature is the first geometric feature or another second geometric feature which is determined and is different from the second geometric feature to be determined.

In this embodiment, the first geometric feature is determined according to the geometric shape of the target contour data segment, one target contour data segment corresponds to one first geometric feature, and the second geometric feature may be determined according to the geometric relationship between two first geometric features, may be determined according to the geometric relationship between one first geometric feature and another determined second geometric feature, or may be determined according to the geometric relationship between two determined second geometric features, that is, the two reference geometric features may be the first geometric feature, the determined second geometric feature, the determined first geometric feature, and the determined second geometric feature. Geometric shapes include, but are not limited to, straight lines, circular arcs, and points. Geometric relationships include, but are not limited to, intersecting, passing points parallel to a straight line, straight lines defined by two points, symmetry, and the like.

Step S104, selecting target geometric features meeting the visual measurement requirements of the item to be measured from the geometric features.

In this embodiment, the target geometric feature may be one or more, may be a first geometric feature, may be a second geometric feature, and may also be the first geometric feature and the second geometric feature, which are specifically determined according to the measurement requirement of the item to be measured.

Step S105, calculating the measurement data of the item to be measured according to the target geometric features and the measurement strategy.

In this embodiment, as an implementation manner, the measurement policy may be abstracted in advance and implemented independently, so as to achieve separation and decoupling of the measurement policy and the input data, at this time, the target geometric feature is used as input of the measurement policy, the measurement data of the items to be measured may be directly obtained according to the measurement policy, different items to be measured may adopt the same measurement policy, but the target geometric feature of the input measurement policy is different, as another implementation manner, each item to be measured may also correspond to one measurement policy, at this time, when different items to be measured correspond to the same measurement policy, two maintenance units are required to implement the same measurement policy, and maintenance cost is high.

According to the method provided by the embodiment, the first geometric feature or the first geometric feature and the second geometric feature which meet the measurement requirement are automatically extracted, and the first geometric feature characterizes the geometric form of the target outline data segment, and the second geometric feature characterizes the geometric relation of the target outline data segment, so that the measurement data of the item to be measured can be directly calculated according to the target geometric feature and the measurement strategy selected from the first geometric feature and the second geometric feature, and the process of reprocessing the directly calculated measurement data when the measurement data cannot meet the measurement requirement is avoided, so that the measurement efficiency is improved.

In an alternative embodiment, each target profile data segment corresponds to a first extraction policy that meets the visual measurement requirement of the item to be measured, according to different measurement requirements of different items to be measured, different target profile data segments may correspond to the same first extraction policy or may correspond to different first extraction policies, the same target profile data segment may correspond to the same first extraction policy or may correspond to different first extraction policies for different items to be measured, for example, for the same target profile data segment, the measurement requirement a needs to extract a straight line corresponding to the target profile data segment, and the measurement requirement B needs to extract a feature point of the target profile data segment.

The first extraction strategy is used for extracting the first geometric feature, and as an implementation manner, the embodiment can abstract the implementation of the first extraction strategy, realize the separation and the decoupling of the first extraction strategy and the target contour data segment, maximally multiplex the first extraction strategy, and for any contour data segment to be extracted in the target contour data segment, one implementation manner for extracting the first geometric feature is as follows:

if the first extraction strategy is feature point extraction, taking feature points of the contour point cloud in the contour data segment to be extracted as first geometric features;

And if the first extraction strategy is geometric fitting, fitting all contour point clouds of the contour data segment to be extracted according to the geometric form of the contour data segment to be extracted to obtain a first geometric feature.

In the present embodiment, the feature points include, but are not limited to, extreme points, centroid points, and nearest points. Taking an extreme point as an example, the extreme point comprises a maximum value and a minimum value, and for the contour point cloud in the contour data segment to be extractedIn the sense that the number of the cells,To mention the first%Cloud of outline points,/>Representing the number of contour point clouds in a contour data segment to be extracted,/>And/>Respectively represent the/>/>, Of the outline Point cloudAxis coordinate sum/>The axis coordinates can specify whether the maximum value or the minimum value is taken or the main direction characteristic, namely the maximum value or the minimum value is aimed at/>, according to the requirement of the measured dataThe axis coordinates are also for/>And (5) axis coordinates. Taking a centroid point as an example, if the contour data segment to be extracted is an arc, the centroid point may be the centroid of the arc. Taking the nearest point as an example, for a designated preset point, extracting a point closest to the preset point in the contour point cloud in the contour data segment to be extracted as a characteristic point.

In this embodiment, geometric fitting includes, but is not limited to, straight line fitting and circular arc fitting, and the adopted fitting mode is to determine an error equation first, and then determine parameters of the error equation according to the contour point cloud, and one implementation mode is as follows:

firstly, determining an error equation of a contour data segment to be extracted according to the geometric form of the contour data segment to be extracted;

Secondly, determining geometric parameters of an error equation according to all contour point clouds of the contour data segment to be extracted;

and finally, taking the error equation after the geometric parameter determination as a first geometric feature.

In this embodiment, taking straight line fitting as an example, an error equation is established for the contour data segment to be extracted: Wherein/> Error equation for straight line,/>，/>Is a straight line parameter,/>Representing the number of contour point clouds in a contour data segment to be extracted,/>And/>Respectively represent the/>/>, Contour point clouds in each contour data segment to be extractedAxis coordinate sum/>The axis coordinates are used for calculating the linear parameter/>, through a least square method or a RANSAC algorithm，/>And obtaining the first geometric feature of the contour data segment to be extracted. Taking circular arc fitting as an example, the error equation is suggested:，/> Error equation for circular arc,/> ，/>，/>Is a circular arc parameter,/>Representing the number of contour point clouds in a contour data segment to be extracted,/>And/>Respectively represent the/>/>, Contour point clouds in each contour data segment to be extractedAxis coordinate sum/>Shaft coordinates, solving the arc parameters/>，/>，/>And obtaining the first geometric feature of the contour data segment to be extracted.

In addition to extracting the first geometric feature, the present embodiment also provides an implementation manner of extracting the second geometric feature:

Firstly, obtaining an extraction configuration meeting the visual measurement requirement of a term to be measured, wherein the extraction configuration comprises a second extraction strategy, a first reference geometric feature and a second reference geometric feature;

second, a second geometric feature is extracted from the first reference geometric feature and the second reference geometric feature.

In this embodiment, for different items to be measured, the first reference geometric feature and the second reference geometric feature may be the same, but the corresponding second extraction strategies may be different, or the first reference geometric feature and the second reference geometric feature may be different, but the corresponding second extraction strategies may be the same, or even different items to be measured, and the extraction configurations are the same, and as an implementation manner, the extraction configuration and the input reference geometric feature may be separated and decoupled, so that the different items to be measured multiplex the same extraction configuration.

In this embodiment, in order to abstract the second extraction policy, the maintenance cost of the second extraction policy is simplified, and this embodiment abstracts the second extraction policy into two types: the intersection point extraction and the straight line extraction are different in the implementation of extracting the second geometric feature for different extraction modes, and one mode is specifically as follows:

Calculating an intersection point of the first reference geometric feature and the second reference geometric feature, and taking the calculated intersection point as the second geometric feature;

in this embodiment, the intersection point may be an intersection point of a straight line and a straight line, or an intersection point of a straight line and an arc, and the intersection point of a straight line and an arc may be one or two.

In this embodiment, for the intersection point of the straight line and the straight line, the calculation method may be:

the first reference geometric feature and the second reference geometric feature are respectively: And/> Wherein/>，/>，/>，/>，/>，/>Are all constant parameters, and calculate the intersection point/>And takes this as the second geometric feature.

For the intersection point of the straight line and the circular arc, the calculation mode can be as follows:

For the first reference geometric feature And a second reference geometrical feature/>Wherein/>，/>，/>，/>，/>，/>Are all constant parameters, let/>,/>,/>，/>Calculate the intersection/>And takes this as the second geometric feature.

For other application scenarios besides rail profile, there may be an intersection point of an arc and an arc, and the intersection point of the arc and the arc may be one or more.

For the straight line extraction method, one method is specifically:

a feature line determined from the first reference geometric feature and the second reference geometric feature is calculated, and the feature line is taken as the second geometric feature.

In this embodiment, the geometric shapes and geometric relationships of the first reference geometric feature and the second reference geometric feature are different, and the manner of determining the feature straight line is also different, and for the case that the first reference geometric feature is a point, the second reference geometric feature is a straight line, and the first reference geometric feature is not on the second reference geometric feature, the determination manner is that:

a straight line passing through the first reference geometric feature and parallel to the second reference geometric feature is determined as a feature straight line.

The manner in which the straight line that passes through the first reference geometric feature and is parallel to the second reference geometric feature may be: for the first reference geometric featureUsing the slope/>, given the second reference geometryCalculate/>As a characteristic straight line.

For the case that the first reference geometric feature and the second reference geometric feature belong to a symmetrical geometric relationship, the determination mode is as follows:

The center line of symmetry of the first reference geometric feature and the second reference geometric feature is determined as a feature straight line.

For the case where the first reference geometric feature is a point and the second reference geometric feature is a point, it is determined in the following manner:

A straight line of the first reference geometric feature and the second reference geometric feature is determined as a feature straight line.

The manner of determining the straight line of the first reference geometry and the second reference geometry may be: for the first reference geometric feature and the second reference geometric featureUsing the formula/>A two-point straight line was obtained, and this was used as a characteristic straight line.

In this embodiment, the measurement strategies include distance measurement and angle measurement, and the calculation modes of the measurement strategies are different for different measurement strategies, and for the distance measurement, the calculation modes are as follows:

and calculating the distance between the two target geometric features, and taking the calculated distance as measurement data.

In this embodiment, the distance between two target geometric features includes, but is not limited to, the distance between two points, the distance from a point to a straight line, the distance from a point to an arc, and the distance from a point to a center of a circle.

When the distance between two target geometric features is the distance between two points, the calculation mode is as follows: for two points, the coordinates are: According to the formula/> Calculate the distance between two points/>As a measurement item measurement.

When the distance between two target geometric features is a point-to-straight line distance, the calculation mode is as follows: for point featuresAnd straight line feature/>Wherein/>，/>，/>Are all constant parameters according to/>Calculate the distance/>As a measurement item measurement.

When the distance between two target geometric features is the point-to-arc distance, the calculation method comprises the following steps: for point featuresAnd arc feature/>Wherein/>，/>，/>，/>Are all constant parameters, satisfying/>Under the condition of-According toCalculate the distance/>As a measurement item measurement.

For angle measurement, the calculation method is as follows:

and calculating an included angle between the two target geometric features, and taking the calculated angle as measurement data.

In this embodiment, the angle between the two target geometric features may be the angle between two straight lines. For example, for straight line features，/>Wherein/>，/>，/>，/>，/>，/>All are constant parameters, calculateAccording to/>Calculating the included angle/>As a measurement value of the measurement item.

It should be noted that, for a relatively complex measurement requirement, the measurement requirement may be decomposed into a plurality of items to be measured, then the plurality of items to be measured are sequentially measured according to a preset sequence, and the measurement results are sequentially combined, so as to finally obtain measurement data of the complex measurement requirement.

In order to more clearly explain the rail measurement method provided by the embodiment, the embodiment provides a measurement process of the rail height of the rail as an item to be measured. Referring to fig. 3, fig. 3 is an exemplary diagram of a rail height measurement process of a rail according to the present embodiment. In fig. 3, the item to be measured is the measurement track height, the measurement strategy is the calculation distance, the selected target profile data segment is a profile data segment 1 composed of profile labels 4 and 17, a profile data segment 2 composed of profile labels 18 and 25, a profile data segment 3 composed of profile label 0, a profile data segment 4 composed of profile label 9, wherein the profile data segment 1 and the profile data segment 2 are straight lines, the first geometric features extracted from the profile data segment 1 and the profile data segment 2 are straight lines 1 and 2 respectively, the profile data segment 3 and the profile data segment 4 are curves, the first geometric features extracted from the profile data segment 3 and the profile data segment 4 are curves 1 and 2 respectively, and the extraction configuration comprises: the contour data segment 3 and the contour data segment 4 are configured as a first reference geometric feature and a second reference geometric feature, the second extraction strategy is configured as straight line extraction, and since the two belong to a symmetrical geometric relationship, the extracted second geometric feature is the symmetrical center line of the two, and the extraction configuration further comprises: respectively configuring a straight line 1 and a symmetrical central line as a first reference geometric feature and a second reference geometric feature, configuring a second extraction strategy as intersection point extraction, and taking an intersection point b of the straight line 1 and the symmetrical central line as a second geometric feature; the extraction configuration further includes: the straight line 2 and the symmetrical center line are respectively configured as a first reference geometric feature and a second reference geometric feature, the second extraction strategy is configured as intersection point extraction, and an intersection point a of the two is used as the second geometric feature. Thus, the extracted geometric features include: straight line 1, straight line 2, symmetrical central line, crossing point a and crossing point b, choose crossing point a and crossing point b as the goal geometric feature from among them, calculate the distance between two according to the measurement policy, regard calculated distance as the orbit height.

The embodiment also provides a measuring process of the track top width of the item to be measured. Referring to fig. 4, fig. 4 is an exemplary diagram of a rail height measurement process of a rail according to the present embodiment. In fig. 4, the item to be measured is the top width of the measurement track, the measurement strategy is the calculation distance, the selected target profile data segment is the profile data segment 1 composed of profile labels 4 and 17, the profile data segment 2 composed of profile labels 21, and the profile data segment 3 composed of profile labels 23, then the first geometric features extracted from the profile data segment 1, the profile data segment 2 and the profile data segment 3 are straight line 1, straight line 2 and straight line 3, respectively, and according to the requirement of the item to be measured, the extraction configuration includes: the method comprises the steps of configuring a straight line 1 and a straight line 2 as a first reference geometric feature and a second reference geometric feature respectively, configuring a second extraction strategy as intersection point extraction, configuring an intersection point a of the straight line 1 and the straight line 2 as a second geometric feature, configuring a straight line 1 and a straight line 3 as the first reference geometric feature and the second reference geometric feature respectively, configuring the second extraction strategy as intersection point extraction, and configuring an intersection point b of the straight line 1 and the straight line 3 as the second geometric feature, wherein the extracted geometric features comprise: straight line 1, straight line 2, straight line 3, intersection point a and intersection point b are selected from the points a and b as target geometric features, the distance between the two is calculated according to a measurement strategy, and the calculated distance is used as the track top width.

According to the example provided in the above embodiment, in combination with the rail measurement method provided in the present embodiment, a person skilled in the art can derive other measurement data for the rail, such as measurement data of the rail bottom width, the rail web thickness and the arc of each portion of the rail, without performing creative work.

In the field of rail measurement, after the rail profile data registered with the CAD model is obtained, one implementation method may be: and designing a measuring method directly according to the measured project, and then carrying out a measuring flow of contour segment extraction, geometric feature extraction and contour measurement. In the method for linearly designing the measurement method, for each measurement item, an independent measurement flow is required to be customized and designed, and each measurement item is required to independently perform contour extraction, geometric feature extraction and contour measurement. Wherein at least the following problems exist:

(1) During actual measurement, the contour data segments and the geometric features of each measurement item are isolated from each other and cannot be reused, so that the same contour segment can be repeatedly extracted, the memory load of a hardware platform is increased, and meanwhile, the geometric feature repeated extraction causes the waste of calculation resources.

(2) The measurement system obtained by adopting the linear design measurement mode has poor expandability, and when the measurement project is newly added to the steel rail of the same type or the measurement method is required to be redesigned when the measurement method is replaced by the steel rail of a different type, the measurement system is constructed according to the measurement project, and the development efficiency of the measurement algorithm is seriously influenced. The quality detection link in the steel rail processing process generally requires that the detection system has real-time performance, and when the steel rail profile measurement is carried out, the corresponding measurement algorithm of the measurement system also needs to be capable of measuring the profile section of the steel rail in real time, and the linear design measurement mode obviously cannot meet the real-time measurement requirement.

(3) Line lasers are often used for high-rate data acquisition, the frame rate generally reaches 1000-10000 Hz, and for calculation performance requirements and strictness, for example, rail bottom straight line fitting can be obtained when a rail height is measured, rail bottom concavity also needs to be obtained through rail bottom straight line fitting, rail foot thickness also needs to be obtained through rail bottom straight line fitting, and in addition, due to the fact that a plurality of line laser sensors work simultaneously, each frame of data is huge. The limitation of the factors greatly influences the application and the measurement efficiency of the linear measurement system in an actual steel rail measurement scene, and severely restricts the steel rail profile measurement efficiency.

In view of this, this embodiment combines above-mentioned rail measurement method, abstracts the rail measurement process that this embodiment provided, decouples each stage, and the convenience multiplexes the processing procedure of each stage, further improves measurement efficiency, and this embodiment, its main improvement thinking lies in: abstract the rail measurement process into 3 kinds of nodes: the data nodes, the feature nodes and the measurement nodes can be multiple, each feature node corresponds to different feature extraction strategies, and each measurement node corresponds to different measurement strategies. And selecting required characteristic nodes and measuring nodes according to the measurement requirements of the items to be measured, organizing the data nodes, the selected characteristic nodes and the measuring nodes according to the graph model according to the measurement requirements, wherein each node corresponds to one vertex of the graph model, and the input-output relationship between the two nodes is an edge between the corresponding vertices of the graph model. According to the measurement requirement of the item to be measured, the profile point cloud data of the corresponding steel rail with the profile label can be configured for the data node to serve as the input of the data node, the data node outputs the corresponding profile data segment according to the input, the selected feature node takes the output of the data node as the input of the data node, the corresponding geometric feature is output, the selected measurement node selects the required target geometric feature from the geometric features output by the feature node to serve as the input, and the measurement result is output. And the high-efficiency measurement of different steels can be completed by only one set of codes and simple configuration.

It should be noted that, in order to further improve the multiplexing rate and finally improve the measurement efficiency, each feature node may store the geometric feature of the target profile data segment extracted by the present node, so that when the next measurement requirement needs to extract the same geometric feature for the target profile data segment, the stored geometric feature is directly used without repeated extraction, for example, the item to be measured a needs to extract the maximum value of the profile data segment 1, and for this purpose, feature node a is configured, and after the feature node a extracts the maximum value of the profile data segment 1, the maximum value of the profile data segment is stored. Then, the item B to be measured also needs to carry out maximum extraction on the profile data segment 1, therefore, the feature node a is configured, the feature node a is already extracted before finding the maximum value of the profile data segment 1 when carrying out geometric feature extraction, and the maximum value of the profile data segment 1 stored by the feature node can be directly output, so that the extraction calculation amount of the maximum value of the profile data segment 1 is reduced.

Based on the above-mentioned thought, the present embodiment provides a rail measurement system, where the rail measurement system includes a plurality of nodes organized based on a graph model according to an item to be measured, the plurality of nodes include data nodes, feature nodes, and measurement nodes, each node is a vertex of the graph model, and an input-output relationship between two nodes is an edge between corresponding vertices of the graph model;

The data node is used for acquiring the profile data segments and the items to be measured of the steel rail to be measured, selecting a plurality of target profile data segments from the profile data segments according to the items to be measured, inputting the plurality of target profile data segments into the characteristic node, wherein the profile data segments are obtained by dividing the profile point cloud of the steel rail to be measured, and the items to be measured correspond to the measurement strategies;

A feature node for extracting geometric features from the multiple segments of target profile data segments and selecting target geometric features from the geometric features that meet the visual measurement requirements of the item to be measured and inputting the target geometric features to the measurement node, the geometric features including a first geometric feature of each target profile data segment or including a first geometric feature of each target profile data segment and at least one second geometric feature, each first geometric feature being determined by the geometry of the corresponding target profile data segment, each second geometric feature being determined by the geometric relationship of two reference geometric features, each reference geometric feature being either a first geometric feature or a determined and different second geometric feature than the second geometric feature that needs to be determined;

and the measurement node is used for calculating measurement data of the item to be measured according to the target geometric characteristics and the measurement strategy.

In an alternative embodiment, each target profile data segment corresponds to a first extraction strategy that meets the visual measurement requirement of the item to be measured, and the feature node is specifically configured to:

For any contour data segment to be extracted in the target contour data segment, if the first extraction strategy is feature point extraction, taking feature points of contour point clouds in the contour data segment to be extracted as first geometric features;

In an optional embodiment, the feature node is configured to fit all contour point clouds of the to-be-extracted contour data segment according to the geometry of the to-be-extracted contour data segment, and when obtaining the first geometric feature, the feature node is specifically configured to:

determining geometric parameters of an error equation according to all contour point clouds of the contour data segment to be extracted;

And taking the error equation after the geometric parameter determination as a first geometric feature.

In an alternative embodiment, the feature node is further configured to:

acquiring an extraction configuration meeting the visual measurement requirement of a term to be measured, wherein the extraction configuration comprises a second extraction strategy, a first reference geometric feature and a second reference geometric feature;

The second geometric feature is extracted from the first reference geometric feature and the second reference geometric feature.

In an alternative embodiment, the feature node is specifically configured to, when configured to extract a second geometric feature from the first reference geometric feature and the second reference geometric feature:

In an alternative embodiment, the feature node is specifically adapted when used for calculating a feature line determined by the first reference geometrical feature and the second reference geometrical feature:

If the first reference geometric feature is a point, the second reference geometric feature is a straight line, and the first reference geometric feature is not positioned on the second reference geometric feature, determining the straight line passing through the first reference geometric feature and parallel to the second reference geometric feature as a feature straight line;

If the first reference geometric feature and the second reference geometric feature belong to a symmetrical geometric relationship, determining the symmetrical center lines of the first reference geometric feature and the second reference geometric feature as feature straight lines;

If the first reference geometric feature is a point and the second reference geometric feature is a point, determining a straight line of the first reference geometric feature and the second reference geometric feature as a feature straight line.

In an alternative embodiment, the target geometry is two, the measuring node being specifically adapted to:

According to the rail measurement system composed of the plurality of nodes organized based on the graph model, the reusability of feature extraction and measurement calculation is greatly improved, for different measurement requirements of different rails, efficient measurement of different measurement requirements of different rails can be completed by only realizing one set of feature extraction and measurement calculation codes, selecting corresponding feature nodes and measurement nodes and simply configuring the corresponding feature nodes and the measurement nodes, and the measurement efficiency is improved.

In order to more specifically describe the operation process of the rail measurement system, please refer to fig. 5, fig. 5 is an exemplary diagram of an overall rail measurement architecture provided in the present embodiment, fig. 5 is:

The input of the data node is the contour point cloud of the steel rail with the contour label and the measurement requirement of the item to be measured, and the output is the target contour data segment meeting the item to be measured, so that the decoupling between the item to be measured and the contour data segment is realized, and for the same steel rail, different items to be measured can select the target contour data segment meeting the requirement of the user from the set of the contour point cloud of the steel rail with the contour label.

The input of the feature nodes is a target contour data segment and an extraction strategy, the output is a geometric feature, and the geometric feature comprises a first geometric feature and a second geometric feature, so that the processing of extracting the geometric feature is decoupled, and the multiplexing of the same geometric feature extraction process is realized.

The input of the measuring node is the geometric feature, the item to be measured and the measuring strategy, and the output is the measuring result. The measurement node selects target geometric features from geometric features according to items to be measured, calculates the target geometric features according to a measurement strategy to obtain a measurement result, so that the processing of the measurement strategy is decoupled, and the multiplexing of the calculation process of the same measurement strategy is realized.

The data nodes, the characteristic nodes and the measuring nodes in the figure 5 correspond to the data layer, the characteristic layer and the measuring layer respectively, when the profile of the steel rail is measured, the connection condition of the nodes of different levels is utilized to construct the data flow of the nodes of the figure, and different measuring algorithm processes are realized based on data flow transmission, so that the measurement of different measuring items of the steel rail is realized, repeated copying of data in the calculation process is reduced, the geometric features extracted from the measuring items with the same geometric features have reusability, and the measuring efficiency of the steel rail is effectively improved; meanwhile, by utilizing an algorithm frame platform and intelligently organizing different nodes, the flow control of different measurement projects is realized, and the development efficiency of different steel rail measurement algorithms is improved.

In order to carry out the respective steps of the above-described examples and of the various possible embodiments, an implementation of the rail measuring device 100 is given below. Referring to fig. 6, fig. 6 is a block schematic diagram of a rail measuring device provided by the present invention, and it should be noted that the basic principle and the technical effects of the rail measuring device 100 provided by the present invention are the same as those of the corresponding embodiment, and for brevity, the description of this embodiment is not mentioned.

The rail measurement device 100 includes an acquisition module 110, a selection module 120, an extraction module 130, and a calculation module 140.

The acquisition module 110 is configured to acquire a profile data segment of a rail to be measured and a measurement item, where the profile data segment is obtained by dividing a profile point cloud of the rail to be measured, and the measurement item corresponds to a measurement policy;

a selecting module 120, configured to select a multi-segment target contour data segment from the contour data segments according to the item to be measured;

An extraction module 130, configured to extract geometric features from the multiple segments of target profile data segments, where the geometric features include a first geometric feature of each target profile data segment or include a first geometric feature of each target profile data segment and at least one second geometric feature, where each first geometric feature is determined by a geometric shape of a corresponding target profile data segment, and each second geometric feature is determined by a geometric relationship between two reference geometric features, and each reference geometric feature is a first geometric feature or a determined second geometric feature different from the second geometric feature that needs to be determined;

The selection module 120 is further configured to select a target geometric feature that meets the visual measurement requirement of the item to be measured from the geometric features;

the calculation module 140 is configured to calculate measurement data of the item to be measured according to the target geometric feature and the measurement strategy.

In an alternative embodiment, each target profile data segment corresponds to a first extraction policy that meets the visual measurement requirement of the item to be measured, and the extraction module 130 is specifically configured to: for any contour data segment to be extracted in the target contour data segment, if the first extraction strategy is feature point extraction, taking feature points of contour point clouds in the contour data segment to be extracted as first geometric features; and if the first extraction strategy is geometric fitting, fitting all contour point clouds of the contour data segment to be extracted according to the geometric form of the contour data segment to be extracted to obtain a first geometric feature.

In an alternative embodiment, the extracting module 130 is configured to fit all the contour point clouds of the contour data segment to be extracted according to the geometry of the contour data segment to be extracted, so as to obtain the first geometric feature, and specifically is configured to: determining an error equation of the profile data segment to be extracted according to the geometric form of the profile data segment to be extracted; determining geometric parameters of an error equation according to all contour point clouds of the contour data segment to be extracted; and taking the error equation after the geometric parameter determination as a first geometric feature.

In an alternative embodiment, the extracting module 130 is specifically further configured to: acquiring an extraction configuration meeting the visual measurement requirement of a term to be measured, wherein the extraction configuration comprises a second extraction strategy, a first reference geometric feature and a second reference geometric feature; the second geometric feature is extracted from the first reference geometric feature and the second reference geometric feature.

In an alternative embodiment, the extraction module 130 is specifically configured to, when configured to extract the second geometric feature from the first reference geometric feature and the second reference geometric feature: if the second extraction strategy is intersection point extraction, calculating intersection points of the first reference geometric features and the second reference geometric features, and taking the calculated intersection points as the second geometric features; if the second extraction strategy is linear extraction, calculating a characteristic straight line determined by the first reference geometric feature and the second reference geometric feature, and taking the characteristic straight line as the second geometric feature.

In an alternative embodiment, the extraction module 130 is specifically configured to, when configured to calculate the feature straight line determined by the first reference geometric feature and the second reference geometric feature: if the first reference geometric feature is a point, the second reference geometric feature is a straight line, and the first reference geometric feature is not positioned on the second reference geometric feature, determining the straight line passing through the first reference geometric feature and parallel to the second reference geometric feature as a feature straight line; if the first reference geometric feature and the second reference geometric feature belong to a symmetrical geometric relationship, determining the symmetrical center lines of the first reference geometric feature and the second reference geometric feature as feature straight lines; if the first reference geometric feature is a point and the second reference geometric feature is a point, determining a straight line of the first reference geometric feature and the second reference geometric feature as a feature straight line.

In an alternative embodiment, the target geometry is two, and the computing module 140 is specifically configured to: if the measurement strategy is distance measurement, calculating the distance between the two target geometric features, and taking the calculated distance as measurement data; if the measurement strategy is angle measurement, calculating an included angle between the two target geometric features, and taking the calculated angle as measurement data.

Referring to fig. 7, fig. 7 shows a schematic block diagram of the electronic device 10, where the electronic device 10 includes a processor 11, a memory 12 and a bus 13, and the processor 11 and the memory 12 are connected through the bus 13.

The processor 11 may be an integrated circuit chip with signal processing capabilities. In practice, the steps of the rail measurement method of the above embodiment may be performed by instructions in the form of hardware integrated logic circuits or software in the processor 11. The processor 11 may be a general-purpose processor including a CPU (Central Processing Unit ), NP (Network Processor, network processor), and the like; but also DSP (DIGITAL SIGNAL Processor), ASIC (Application SPECIFIC INTEGRATED Circuit), FPGA (Field Programmable Logic GATE ARRAY, field programmable gate array) or other programmable logic device, discrete gate or transistor logic device, discrete hardware components.

The memory 12 is used to store a program for implementing the rail measurement method of the above-described embodiment, and the program may be a software function module stored in the memory 12 in the form of software or firmware (firmware) or solidified in an OS (Operating System) of the electronic device 10. After receiving the execution instruction, the processor 11 executes a program to implement the rail measurement method disclosed in the above embodiment.

An embodiment of the present invention provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements a rail measurement method according to any one of the preceding embodiments.

In summary, the embodiment of the invention provides a method, a device, an electronic device, a readable storage medium and a system for measuring a steel rail, wherein the method comprises the following steps: acquiring a profile data segment of a steel rail to be measured and a project to be measured, wherein the profile data segment is obtained by dividing a profile point cloud of the steel rail to be measured, and the project to be measured corresponds to a measurement strategy; selecting a plurality of target contour data segments from the contour data segments according to the item to be measured; extracting geometric features from the multiple segments of target contour data segments, the geometric features comprising a first geometric feature of each target contour data segment or comprising a first geometric feature of each target contour data segment and at least one second geometric feature, each first geometric feature being determined by the geometry of the corresponding target contour data segment, each second geometric feature being determined by the geometric relationship of two reference geometric features, each reference geometric feature being a first geometric feature or another second geometric feature that has been determined and that is different from the second geometric feature that needs to be determined; selecting target geometric features meeting the visual measurement requirements of items to be measured from the geometric features; and calculating measurement data of the item to be measured according to the target geometric features and the measurement strategy. Compared with the prior art, the embodiment has at least the following advantages: (1) By automatically extracting the first geometric feature or the first geometric feature and the second geometric feature which meet the measurement requirement, the geometric shape of the target outline data segment is represented by the first geometric feature, the geometric relationship of the target outline data segment is represented by the second geometric feature, and the measurement data of the item to be measured can be directly calculated according to the target geometric feature and the measurement strategy selected from the geometric relationship, so that the process of reprocessing the directly calculated measurement data when the measurement requirement cannot be met is avoided, and the measurement efficiency is improved. (2) The connection condition of the nodes of different levels is utilized to construct the data flow of the graph nodes, and different measurement algorithm processes are realized based on data flow transmission, so that the measurement of different measurement items of the steel rail is realized, repeated copying of data in the calculation process is reduced, the geometric features extracted from the measurement items with the same geometric features have reusability, and the measurement efficiency of the steel rail is effectively improved; meanwhile, the method in the embodiment is realized in a frame platform mode, different nodes can be intelligently organized, flow control of different measurement projects is realized, and development efficiency of different steel rail measurement algorithms is improved.

The foregoing is merely illustrative of the present invention, and the present invention is not limited thereto, and any changes or substitutions easily contemplated by those skilled in the art within the scope of the present invention should be included in the present invention. Therefore, the protection scope of the invention is subject to the protection scope of the claims.

Claims

1. A method of rail measurement, the method comprising:

2. A rail measurement method as claimed in claim 1 wherein each of said target profile data segments corresponds to a first extraction strategy that meets the visual measurement requirements of said item to be measured, said step of extracting geometric features from said plurality of segments of target profile data segments comprising:

3. The method for measuring steel rails as in claim 2 wherein said step of fitting all contour point clouds of said contour data segment to be extracted according to the geometry of said contour data segment to be extracted to obtain said first geometric feature comprises:

4. A rail measurement method as claimed in claim 1 wherein the step of extracting geometric features from the multi-segment target profile data segment further comprises:

5. A rail measurement method as claimed in claim 4 wherein the step of extracting a second geometric feature from the first and second reference geometric features comprises:

6. A rail measurement method as claimed in claim 5 wherein the step of calculating a characteristic line defined by the first reference geometry and the second reference geometry comprises:

7. A rail measurement method as claimed in claim 1, wherein the target geometry is two, and the step of calculating measurement data of the item to be measured based on the target geometry and the measurement strategy comprises:

8. A rail measurement device, the device comprising:

9. An electronic device comprising a processor and a memory, the memory for storing a program, the processor for implementing the rail measurement method of any one of claims 1-7 when the program is executed.

10. A computer readable storage medium, characterized in that a computer program is stored thereon, which computer program, when being executed by a processor, implements a rail measurement method as claimed in any one of claims 1-7.

11. The steel rail measurement system is characterized by comprising a plurality of nodes organized based on a graph model according to items to be measured, wherein the nodes comprise data nodes, characteristic nodes and measurement nodes, each node is a vertex of the graph model, and the input-output relationship between the two nodes is an edge between corresponding vertices of the graph model;