CN115830019A

CN115830019A - Three-dimensional point cloud calibration processing method and device for steel rail detection

Info

Publication number: CN115830019A
Application number: CN202310109398.1A
Authority: CN
Inventors: 黄碗明; 鲁鑫; 黄汝成
Original assignee: Nanjing Huiran Technology Co ltd
Current assignee: Nanjing Huiran Technology Co ltd
Priority date: 2023-02-14
Filing date: 2023-02-14
Publication date: 2023-03-21
Anticipated expiration: 2043-02-14
Also published as: CN115830019B

Abstract

The invention provides a three-dimensional point cloud calibration processing method and a three-dimensional point cloud calibration processing device for steel rail detection, wherein point cloud three-dimensional images of positions of steel rails are collected at preset heights and preset positions by a point cloud collection device, and three-dimensional coordinates of all points in the point cloud three-dimensional images are determined; calculating according to the three-dimensional coordinates of the equipment, the distance information between the steel rails and the width information of the steel rails to obtain a target coordinate acquisition interval; determining a corresponding target area in the point cloud three-dimensional image as a steel rail detection calibration area according to the target coordinate acquisition interval, and performing detection judgment on all point three-dimensional coordinates in the steel rail detection calibration area according to standard detection information to determine an abnormal area in the steel rail detection calibration area; and generating a corresponding processing display frame according to the attribute of the three-dimensional coordinates of the points in the abnormal area, and combining the processing display frame with the abnormal area to display in the point cloud three-dimensional image and generate a processing result.

Description

Three-dimensional point cloud calibration processing method and device for steel rail detection

Technical Field

The invention relates to the technical field of data processing, in particular to a three-dimensional point cloud calibration processing method and device for steel rail detection.

Background

With the high-speed development of Chinese economic construction, the population is more concentrated by modern high-speed operation capability and urban construction, and the pressure of high-speed rail transportation and urban traffic is rapidly increased, so that the construction of rail transportation facilities is actively developed in China, and at present, high-speed rail, subway and urban rail transportation networks which account for hundreds of thousands of kilometers are built, so that the operation safety of trains is the primary problem of rail transportation. The steel rail is used as the most basic part in the track, and the working state of the steel rail is directly related to the train running safety; the rail detection mainly detects the state of the rail manually by means of detection equipment, and mainly detects the defects of corrosion, cracks, depressions, bulges and the like of the steel rail.

In the prior art, the surface of the steel rail is mainly inspected by manually utilizing the detection equipment, compared with three-dimensional point cloud data, the detection accuracy is not high, and the efficiency is very low when the inspection is manually performed continuously, so that great potential safety hazards exist.

Therefore, how to provide an efficient and high-accuracy automatic steel rail detection method becomes an urgent problem to be solved.

Disclosure of Invention

The embodiment of the invention provides a three-dimensional point cloud calibration method and device for steel rail detection, which can automatically perform de-duplication processing on data outside a steel rail, automatically detect the steel rail by using the three-dimensional point cloud data, automatically display related problems and better improve the accuracy and efficiency.

In a first aspect of the embodiments of the present invention, a three-dimensional point cloud calibration processing method for steel rail detection is provided, including:

controlling a point cloud acquisition device to acquire a point cloud three-dimensional image of the position of the steel rail at a preset height and a preset position, and determining three-dimensional coordinates of all points in the point cloud three-dimensional image;

receiving size data configured by a user for a steel rail, wherein the size data at least comprises steel rail distance information and steel rail width information, and calculating according to the three-dimensional coordinates of equipment, the steel rail distance information and the steel rail width information to obtain a target coordinate acquisition interval;

determining a corresponding target area in the point cloud three-dimensional image as a steel rail detection calibration area according to the target coordinate acquisition interval, and performing detection judgment on all point three-dimensional coordinates in the steel rail detection calibration area according to standard detection information to determine an abnormal area in the steel rail detection calibration area;

and generating a corresponding processing display frame according to the attribute of the three-dimensional coordinates of the points in the abnormal area, and combining the processing display frame and the abnormal area to display in the point cloud three-dimensional image and generate a processing result.

Optionally, in a possible implementation manner of the first aspect, the controlling the point cloud collecting device to collect a point cloud three-dimensional image of a position where the steel rail is located at a preset height and a preset position, and determining three-dimensional coordinates of all points in the point cloud three-dimensional image includes:

controlling the point cloud acquisition equipment to be at a preset height, taking a midpoint between the two steel rails as a preset position, and controlling the point cloud acquisition equipment to downwards acquire a point cloud three-dimensional image corresponding to the steel rails after judging that the point cloud acquisition equipment reaches the preset height and the preset horizontal position;

and constructing a three-dimensional coordinate system by taking the point cloud acquisition equipment as a coordinate origin, and determining the three-dimensional coordinates of all points in the point cloud three-dimensional image.

Optionally, in a possible implementation manner of the first aspect, the receiving size data configured by the user for the steel rail, where the size data at least includes information on a distance between steel rails and information on a width of the steel rail, and performing calculation according to the three-dimensional coordinates of the device, the information on the distance between steel rails, and the information on the width of the steel rail to obtain a target coordinate acquisition interval includes:

calculating a half distance value corresponding to the distance information between the steel rails, and extending the half distance value to the positive side and the negative side of the X axis by taking the horizontal coordinate of the three-dimensional coordinate of the equipment as a starting point to obtain a positive initial X point coordinate and a negative initial X point coordinate;

according to the width information of the steel rail, respectively determining coordinate points in a positive direction and a negative direction by taking a positive starting X point coordinate and a negative starting X point coordinate as starting points to obtain a positive ending X point coordinate and a negative ending X point coordinate;

counting all X-axis coordinate points from the positive starting X-point coordinate to the positive ending X-point coordinate to obtain a positive X-axis coordinate interval, counting all X-axis coordinate points from the negative starting X-point coordinate to the negative ending X-point coordinate to obtain a negative X-axis coordinate interval, and obtaining a target coordinate collection interval according to the positive X-axis coordinate interval and the negative X-axis coordinate interval.

Optionally, in a possible implementation manner of the first aspect, the determining, according to the target coordinate acquisition section, a corresponding target area in the point cloud three-dimensional image as a rail detection calibration area, and performing detection and judgment on all point three-dimensional coordinates in the rail detection calibration area according to standard detection information to determine an abnormal area in the rail detection calibration area includes:

comparing X-axis coordinates in three-dimensional coordinates of all points in the point cloud three-dimensional image with a target coordinate acquisition interval respectively, and determining a point three-dimensional coordinate of which the X-axis coordinate is positioned in the target coordinate acquisition interval as a target point;

counting a target area formed by a target point corresponding to a positive X-axis coordinate as a positive steel rail detection calibration area, and counting a target area formed by a target point corresponding to a negative X-axis coordinate as a negative steel rail detection calibration area;

and calling standard detection information corresponding to the steel rail, determining Z-axis coordinates of all target points in the steel rail detection calibration area, comparing the Z-axis coordinates with the standard detection information to perform detection judgment, and determining an abnormal area in the steel rail detection calibration area.

Optionally, in a possible implementation manner of the first aspect, the calling standard detection information corresponding to a steel rail, determining Z-axis coordinates of all target points in a steel rail detection calibration area, comparing the Z-axis coordinates with the standard detection information to perform detection and judgment, and determining an abnormal area in the steel rail detection calibration area includes:

comparing the Z-axis coordinate of each target point with the standard Z-axis coordinate in the standard detection information to obtain a Z-axis coordinate difference value, and if the Z-axis coordinate difference value is greater than a preset coordinate difference value, judging that the target point corresponding to the corresponding Z-axis coordinate is an abnormal target point;

if the Z-axis coordinate difference is smaller than or equal to the preset coordinate difference, judging that the target point corresponding to the corresponding Z-axis coordinate is a normal target point;

and calculating the distance between any two abnormal target points to obtain the distance between the target points, and classifying all the abnormal target points of which the distances between the target points are smaller than the preset distance between the points into the same abnormal area.

Optionally, in a possible implementation manner of the first aspect, before determining, according to the target coordinate acquisition interval, a corresponding target area in the point cloud three-dimensional image as a rail detection calibration area, performing detection and judgment on three-dimensional coordinates of all points in the rail detection calibration area according to standard detection information, and determining an abnormal area in the rail detection calibration area, the method further includes:

acquiring preset widths of inclined surfaces on two sides of a steel rail, and respectively determining coordinate points in a corresponding positive direction and a corresponding negative direction by taking a positive initial X point coordinate and a negative initial X point coordinate as starting points according to the preset widths to obtain a first positive removal X point coordinate and a first negative removal X point coordinate;

according to the preset width, coordinate point determination is respectively carried out towards a corresponding negative direction and a corresponding positive direction by taking the positive termination X point coordinate and the negative termination X point coordinate as starting points, and a second positive removal X point coordinate and a second negative removal X point coordinate are obtained;

and counting all the X-axis coordinate points of the first positive-direction removed X-point coordinate and the second positive-direction removed X-point coordinate to obtain a positive-direction X-axis coordinate interval, counting all the X-axis coordinate points of the first negative-direction removed X-point coordinate and the second negative-direction removed X-point coordinate to obtain a negative-direction X-axis coordinate interval, and obtaining a target coordinate collection interval according to the positive-direction X-axis coordinate interval and the negative-direction X-axis coordinate interval.

Optionally, in a possible implementation manner of the first aspect, the calculating a distance between any two abnormal target points to obtain a distance between the target points, and classifying all the abnormal target points whose distances between the target points are smaller than a preset distance between the preset points into the same abnormal area includes:

adding different area marks to all the abnormal target points, and calculating the distance between any two abnormal target points to obtain the distance between the target points;

synchronously modifying the area marks of all the abnormal target points with the distance between the target points smaller than the distance between the preset points in a way of copying the information of the area marks of other abnormal target points and adding the information into the information of the corresponding area marks;

if the distance between the target point with one abnormal target point and the other abnormal target points is smaller than the preset distance, synchronously modifying the area marks of the corresponding abnormal target point and the other abnormal target points;

after the area marks of all the abnormal target points are judged to be synchronously modified respectively, the abnormal target points with the same area marks are classified into the same abnormal area.

Optionally, in a possible implementation manner of the first aspect, the generating a corresponding processing display frame according to an attribute of a three-dimensional coordinate of a point in the abnormal region, and combining the processing display frame and the abnormal region to display in a point cloud three-dimensional image and generate a processing result includes:

determining the attribute of a three-dimensional coordinate of a point in each abnormal area, wherein the attribute of the three-dimensional coordinate of the point comprises the number of all abnormal target points and the three-dimensional coordinate of each abnormal target point;

extracting an X-axis extreme value and a Y-axis extreme value in a three-dimensional coordinate of an abnormal target point corresponding to each abnormal area, and determining two groups of X-axis parallel lines and Y-axis parallel lines according to the X-axis extreme value and the Y-axis extreme value;

according to the X-axis extreme value and the Y-axis extreme value, calculating the X-axis distance and the Y-axis distance of the processing display frame, and according to the X-axis distance, the Y-axis distance and the number of the three-dimensional coordinates of all points, so as to obtain the display color of the processing display frame;

determining a central point of the processing display frame and a central point of the abnormal region, and setting the central point of the processing display frame to be overlapped with the central point of the abnormal region so as to enable the processing display frame and the abnormal region to be combined and displayed in the point cloud three-dimensional image and generate a processing result, wherein the processing result comprises the number of point three-dimensional coordinates.

Optionally, in a possible implementation manner of the first aspect, the step of calculating, according to the X-axis distance, the Y-axis distance, and the number of three-dimensional coordinates of all points, the display color of the processing display frame by using a frame line formed according to the two sets of X-axis parallel lines and Y-axis parallel lines as the processing display frame and according to the X-axis extreme value and the Y-axis extreme value as the X-axis distance and the Y-axis distance of the processing display frame includes:

determining a frame line formed by intersection areas of two groups of X-axis parallel lines and Y-axis parallel lines as a processing display frame, and amplifying the processing display frame by a preset multiple;

calculating according to the X-axis extreme value, the Y-axis extreme value and the preset multiple to obtain the X-axis distance and the Y-axis distance of the processing display frame after being amplified, and obtaining the area of the processing display frame according to the X-axis distance and the Y-axis distance after being amplified;

obtaining three-dimensional coordinate density according to the area of the display frame and the number of the three-dimensional coordinates of the points, comparing the three-dimensional coordinate density with preset density intervals to obtain corresponding display colors, wherein each preset density interval has a corresponding display color, calculating the three-dimensional coordinate density by the following formula,

wherein ,

in the form of a three-dimensional coordinate density,

the number of the abnormal target points is,

is the maximum of the extreme values of the X-axis,

is the minimum of the extreme values of the X-axis,

the number of the first time is a preset multiple,

is the maximum of the extreme values of the Y-axis,

the minimum value in the extreme values of the Y axis.

In a second aspect of the embodiments of the present invention, a three-dimensional point cloud calibration processing apparatus for steel rail detection is provided, including:

the control module is used for controlling the point cloud acquisition equipment to acquire a point cloud three-dimensional image of the position of the steel rail at a preset height and a preset position and determining point three-dimensional coordinates of all points in the point cloud three-dimensional image;

the calculation module is used for receiving size data configured by a user for the steel rail, wherein the size data at least comprises steel rail distance information and steel rail width information, and calculating according to the three-dimensional coordinates of the equipment, the steel rail distance information and the steel rail width information to obtain a target coordinate acquisition interval;

the determining module is used for determining a corresponding target area in the point cloud three-dimensional image as a steel rail detection calibration area according to the target coordinate acquisition interval, detecting and judging three-dimensional coordinates of all points in the steel rail detection calibration area according to standard detection information, and determining an abnormal area in the steel rail detection calibration area;

and the generating module is used for generating a corresponding processing display frame according to the attribute of the three-dimensional coordinates of the points in the abnormal area, combining the processing display frame and the abnormal area, displaying the processing display frame and the abnormal area in the point cloud three-dimensional image and generating a processing result.

In a third aspect of the embodiments of the present invention, a storage medium is provided, in which a computer program is stored, which, when being executed by a processor, is adapted to implement the method according to the first aspect of the present invention and various possible designs of the first aspect of the present invention.

The invention provides a three-dimensional point cloud calibration processing method and a device for steel rail detection, which are characterized in that point cloud three-dimensional images at a steel rail are acquired by using point cloud acquisition equipment, a corresponding three-dimensional coordinate system is established in each point cloud three-dimensional image, the point cloud three-dimensional images are subjected to de-weight processing according to the three-dimensional coordinate system, steel rail distance information and steel rail width information to obtain a corresponding target coordinate acquisition interval (steel rail), impurities except the steel rail are automatically removed, and compared with a de-weight mode such as a neural network, the method and the device directly adopt the point cloud acquisition equipment to process through coordinate values, and do not need to train in advance, so that the working efficiency is improved; the three-dimensional coordinate values of all points in the target coordinate acquisition interval are directly compared with the standard in the standard detection information to determine abnormal areas (cracks, pits and the like), corresponding processing display frames are generated in the abnormal areas, and a user can more intuitively determine and position the abnormal areas through the processing display frames.

According to the technical scheme provided by the invention, corresponding point cloud acquisition equipment can be carried by unmanned aerial vehicles and other equipment to acquire a point cloud three-dimensional image in the middle of a steel rail, a coordinate system is established by taking a laser emission point of the point cloud acquisition equipment as a central point, coordinate values of all points in the point cloud three-dimensional image are obtained, a starting point coordinate point (a positive ending X point coordinate and a negative ending X point coordinate) on the inner side of the steel rail is respectively positioned according to distance information between the steel rails, a finishing point coordinate point (a positive ending X point coordinate and a negative ending X point coordinate) on the outer side of the steel rail is determined according to the width information of the steel rail, the positive ending X point coordinate and the negative ending X point coordinate, and a coordinate interval of the upper surface of the steel rail to be acquired is determined according to the starting point coordinate point on the inner side of the steel rail and the finishing point coordinate point on the outer side of the steel rail.

According to the technical scheme provided by the invention, the coordinate values of all coordinate points on the surface of the steel rail, namely the coordinate values of the target point, are determined through the target coordinate acquisition interval, and the difference between the coordinate value of the Z axis in each coordinate value and the coordinate value of the Z axis in the corresponding standard value (standard detection information) is obtained to obtain the Z axis coordinate difference value, so that abnormal areas (recesses, cracks and the like) in the steel rail are determined by comparing the Z axis coordinate difference value with the preset coordinate difference value; if the steel rail has two corresponding inclined surfaces on two sides, the steel rail can be removed by using a coordinate system according to the corresponding preset width of the inclined surfaces, only the plane of the steel rail is reserved, then abnormal areas (pits, cracks and the like) in the steel rail are determined by comparing the coordinate values of the Z axis, the distance between target points of any two points is compared with the distance between preset points, if the distance is smaller than the preset point, the area mark is synchronously modified into the same area mark, and therefore area classification is carried out according to the area mark, and the abnormal areas are automatically positioned.

According to the technical scheme provided by the invention, different processing display frames can be automatically generated according to the different sizes of abnormal regions (such as pits, cracks and the like) to be displayed, so that the convenience is brought to a user to observe, the processing display frames with different colors corresponding to different three-dimensional coordinate densities can be obtained by processing according to the areas of the processing display frames and the number of abnormal target points, and it can be understood that the processing display frames with smaller processing display frames and more abnormal target points have larger corresponding three-dimensional coordinate densities, and the processing display frames with the larger three-dimensional coordinate densities have deeper and more obvious colors corresponding to the larger processing display frames, so that the user can observe the degree and the position of the defect region according to the processing display frames and the corresponding colors conveniently, and the user can perform later maintenance processing conveniently.

Drawings

FIG. 1 is a flow chart of a three-dimensional point cloud calibration processing method for rail detection according to the present invention;

FIG. 2 is a schematic view of an abnormal region of a steel rail according to the present invention;

FIG. 3A is a schematic plan view of a rail according to the present invention;

FIG. 3B is a schematic view of an inclined surface of a rail according to the present invention;

fig. 4 is a schematic structural diagram of a three-dimensional point cloud calibration processing device for rail detection provided by the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

The terms "first," "second," "third," "fourth," and the like in the description and in the claims, as well as in the drawings, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the invention described herein are capable of operation in sequences other than those illustrated or described herein.

It should be understood that, in various embodiments of the present invention, the sequence numbers of the processes do not mean the execution sequence, and the execution sequence of the processes should be determined by the functions and the internal logic of the processes, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

It should be understood that in the present application, "comprising" and "having" and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

It should be understood that, in the present invention, "a plurality" means two or more. "and/or" is merely an association describing an associated object, meaning that three relationships may exist, e.g., a and/or B, may mean: a exists alone, A and B exist simultaneously, and B exists alone. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship. "comprising a, B and C", "comprising a, B, C" means that all three of a, B, C are comprised, "comprising a, B or C" means comprising one of a, B, C, "comprising a, B and/or C" means comprising any 1 or any 2 or 3 of a, B, C.

It should be understood that in the present invention, "B corresponding to a", "a corresponds to B", or "B corresponds to a" means that B is associated with a, and B can be determined from a. Determining B from a does not mean determining B from a alone, but may be determined from a and/or other information. And the matching of A and B means that the similarity of A and B is greater than or equal to a preset threshold value.

As used herein, the term "if" may be interpreted as "at \8230; …" or "in response to a determination" or "in response to a detection" depending on the context.

The technical solution of the present invention will be described in detail below with specific examples. The following several specific embodiments may be combined with each other, and details of the same or similar concepts or processes may not be repeated in some embodiments.

The invention provides a three-dimensional point cloud calibration processing method for steel rail detection, which comprises the following steps of S1 to S5 as shown in figure 1:

s1, controlling point cloud collection equipment to collect a point cloud three-dimensional image of the position where the steel rail is located at a preset height and a preset position, and determining point three-dimensional coordinates of all points in the point cloud three-dimensional image.

The point cloud acquisition equipment can be point cloud acquisition equipment carried by an unmanned aerial vehicle, and the point cloud data is acquired through laser emitted by the point cloud acquisition equipment.

Wherein, predetermine the height that highly for the artificial according to actual conditions predetermines in advance, for example can be 1 meter, 2 meters etc. owing to need detect the defect, predetermine highly should not too high, the precision of gathering is relevant with point cloud collection equipment quality degree, predetermine highly. The preset position is artificially preset in advance according to actual conditions, and the position between two steel rails can be a preset position.

It can be understood that the server controls the point cloud collecting device to collect three-dimensional coordinates of all objects at the steel rail at a preset height and a preset position, so as to obtain a three-dimensional image.

In some embodiments, the step S1 (controlling the point cloud collecting device to collect the point cloud three-dimensional image of the position where the steel rail is located at the preset height and the preset horizontal position, and determining the three-dimensional coordinates of all points in the point cloud three-dimensional image) includes S11 to S12:

and S11, controlling the point cloud acquisition equipment to be at a preset height, taking a midpoint between the two steel rails as a preset position, and controlling the point cloud acquisition equipment to downwards acquire a point cloud three-dimensional image corresponding to the steel rails after judging that the point cloud acquisition equipment reaches the preset height and the preset horizontal position.

It can be understood that, regard the midpoint between two rail as preset position, it should be explained that can connect the intermediate position of slide rail with unmanned through setting up the slide rail between two rail, ensure that the laser emission point is in the rail intermediate position in the cloud of points three-dimensional image that unmanned aerial vehicle gathered, do not limit here.

When the server judges that the point cloud acquisition equipment corresponding to the unmanned aerial vehicle reaches the corresponding preset height and the preset horizontal position, the point cloud acquisition equipment is controlled to acquire the image of the steel rail below, and therefore a point cloud three-dimensional image is obtained.

And S12, constructing a three-dimensional coordinate system by taking the point cloud acquisition equipment as a coordinate origin, and determining the three-dimensional coordinates of all points in the point cloud three-dimensional image.

It can be understood that the point cloud collecting device is provided with a corresponding laser emitting port, the laser emitting port is used as a coordinate origin in the three-dimensional image so as to establish a corresponding three-dimensional coordinate system, and coordinate values of all points in the image can be determined according to the point cloud three-dimensional image and the three-dimensional coordinate system.

For example: the length of the acquisition steel rail set by the point cloud acquisition equipment can be 1 meter, so that three-dimensional images corresponding to 1 meter are obtained, and each three-dimensional image has a coordinate origin corresponding to the three-dimensional image.

And S2, receiving size data configured by a user for the steel rail, wherein the size data at least comprises steel rail distance information and steel rail width information, and calculating according to the three-dimensional coordinates of the equipment, the steel rail distance information and the steel rail width information to obtain a target coordinate acquisition interval.

The distance information between the steel rails is the distance information between the two steel rails, and the width information of the steel rails is the width information of each steel rail; it will be appreciated that both the distance information and the width information between the rails are fixed.

It can be understood that the three-dimensional coordinate of the device is a three-dimensional coordinate system established by establishing a coordinate system with the laser emission port of the device as the origin of coordinates, and the three-dimensional coordinate system is located between two steel rails according to the origin of coordinates, and the distance information between the steel rails and the width information of the steel rails are fixed and can be acquired, so that a corresponding target coordinate acquisition interval (a coordinate interval rail of the steel rails) can be obtained.

Through the mode, other impurities except the steel rail can be removed, such as: the sundries such as stones and plastic bags near the steel rail are only selected from the steel rail, compared with the prior art, a large amount of training is not needed, and the steel rail is automatically detected relatively manually, so that the treatment efficiency is improved.

In some embodiments, the step S2 (receiving the size data configured by the user for the steel rail, where the size data at least includes the distance information between steel rails and the width information of the steel rail, and performing calculation according to the three-dimensional coordinates of the device, the distance information between steel rails, and the width information of the steel rail to obtain the target coordinate collection interval) includes S21 to S23:

and S21, calculating a half distance value corresponding to the distance information between the steel rails, and extending the half distance value to the positive side and the negative side of the X axis by taking the abscissa of the three-dimensional coordinate of the equipment as a starting point to obtain a positive initial X point coordinate and a negative initial X point coordinate.

It should be noted that the origin of coordinates of the three-dimensional coordinate system is located at the middle position of the steel rail, and on the premise that the distance information between the steel rails is known, the corresponding positive-direction starting X-point coordinates and negative-direction starting X-point coordinates (the coordinates of the starting point inside the steel rail) can be determined according to one half of the distance information.

And S22, respectively determining coordinate points in corresponding positive directions and negative directions by taking the positive starting X point coordinate and the negative starting X point coordinate as starting points according to the width information of the steel rail to obtain a positive ending X point coordinate and a negative ending X point coordinate.

It can be understood that after the coordinates of the starting point of the inner side of the steel rail are determined, the coordinates of the positive end point X and the negative end point X (the coordinates of the end point of the outer side of the steel rail) of the outer sides of the two steel rails can be determined according to the inherent steel rail width information of the steel rail.

The corresponding 2 steel rails can be conveniently determined according to the coordinates of the starting point at the inner side of the steel rail and the coordinates of the ending point at the outer side of the steel rail.

S23, counting all X-axis coordinate points from the positive starting X-point coordinate to the positive ending X-point coordinate to obtain a positive X-axis coordinate interval, counting all X-axis coordinate points from the negative starting X-point coordinate to the negative ending X-point coordinate to obtain a negative X-axis coordinate interval, and obtaining a target coordinate collection interval according to the positive X-axis coordinate interval and the negative X-axis coordinate interval.

The positive X-axis coordinate interval is an interval generated by coordinate values of all the X's of the steel rail in the X-axis positive direction of the three-dimensional coordinate system, and the negative X-axis coordinate interval is an interval generated by coordinate values of all the X's of the steel rail in the X-axis negative direction of the three-dimensional coordinate system.

The target coordinate acquisition interval is an interval generated by all X coordinate values of the steel rail in the positive direction and the negative direction of the X axis of the three-dimensional coordinate system.

And S3, determining a corresponding target area in the point cloud three-dimensional image as a steel rail detection calibration area according to the target coordinate acquisition interval, detecting and judging all point three-dimensional coordinates in the steel rail detection calibration area according to standard detection information, and determining an abnormal area in the steel rail detection calibration area.

It can be understood that all coordinate values of the steel rail on all X axes in the three-dimensional coordinate system are known, all three-dimensional coordinate values corresponding to all the coordinate values on the X axes can be determined according to all the coordinate values on the X axes, and the area generated by the three-dimensional coordinate values is used as the steel rail detection calibration area.

By the mode, the coordinate points of the steel rail detection calibration area are compared with all standard coordinate points in the standard detection information, so that the coordinate points with problems are directly positioned.

In some embodiments, before step S3 (determining a corresponding target area in the point cloud three-dimensional image as a rail detection calibration area according to the target coordinate acquisition interval, performing detection and judgment on three-dimensional coordinates of all points in the rail detection calibration area according to standard detection information, and determining an abnormal area in the rail detection calibration area), the method further includes steps A1-A3:

a1, acquiring preset widths of inclined surfaces on two sides of a steel rail, and respectively determining coordinate points in a positive direction and a negative direction by taking a positive initial X point coordinate and a negative initial X point coordinate as starting points according to the preset widths to obtain a first positive removal X point coordinate and a first negative removal X point coordinate.

It should be noted that there are 2 types of rails, one is a rail whose upper surface is a smooth flat surface, and the other is a trapezoidal rail composed of a smooth flat surface and 2 inclined surfaces.

It can be understood that if the steel rail is a trapezoidal steel rail formed by a smooth plane and 2 inclined planes, the preset width, the positive initial X point coordinate and the negative initial X point coordinate of the inclined planes on the two sides of the steel rail are obtained, so that the first positive removal X point coordinate and the first negative removal X point coordinate (the point inside the smooth plane) corresponding to the steel rail are located, and the subsequent point outside the smooth plane can conveniently determine the coordinate set of the upper surface of the steel rail.

And A2, respectively determining coordinate points in a corresponding negative direction and a positive direction by taking the coordinates of the positive termination X point and the negative termination X point as starting points according to the preset width to obtain the coordinates of a second positive removal X point and the coordinates of a second negative removal X point.

It can be understood that, if the steel rail is a trapezoidal steel rail formed by a smooth plane and 2 inclined planes, the two corresponding positive-direction removal X-point coordinates and the second negative-direction removal X-point coordinates (points outside the smooth plane) at the steel rail are located by acquiring the preset width, the positive-direction termination X-point coordinates and the negative-direction termination X-point coordinates of the inclined planes at the two sides of the steel rail.

And A3, counting all X-axis coordinate points of the first positive-direction removed X-point coordinate and the second positive-direction removed X-point coordinate to obtain a positive-direction X-axis coordinate interval, counting all X-axis coordinate points of the first negative-direction removed X-point coordinate and the second negative-direction removed X-point coordinate to obtain a negative-direction X-axis coordinate interval, and obtaining a target coordinate collection interval according to the positive-direction X-axis coordinate interval and the negative-direction X-axis coordinate interval.

It will be appreciated that the coordinate interval of the smooth plane at the upper surface of each rail (target coordinate acquisition interval) may be determined from the points outside the smooth plane and the points inside the smooth plane at each rail.

Through the mode, the inclined surface at the steel rail can be removed.

In some embodiments, the step S3 (determining a corresponding target area in the point cloud three-dimensional image as a rail detection calibration area according to the target coordinate acquisition section, performing detection judgment on all point three-dimensional coordinates in the rail detection calibration area according to standard detection information, and determining an abnormal area in the rail detection calibration area) includes S31 to S33:

and S31, comparing X-axis coordinates in three-dimensional coordinates of all points in the point cloud three-dimensional image with a target coordinate collection interval respectively, and determining the three-dimensional coordinates of the points with the X-axis coordinates in the target coordinate collection interval as target points.

It can be understood that the section generated by all coordinate points on the corresponding X axis can be determined based on the width and the distance of the steel rail, and the point three-dimensional coordinates of all points in the point cloud three-dimensional image of the section generated by all coordinate points on the X axis are determined to be consistent with the point three-dimensional coordinates of all points as the target point.

And S32, counting a target area formed by a target point corresponding to the positive X-axis coordinate as a positive steel rail detection calibration area, and counting a target area formed by a target point corresponding to the negative X-axis coordinate as a negative steel rail detection calibration area.

It can be understood that the X-axis coordinate direction in the three-dimensional coordinate system has both a positive direction and a negative direction; and counting all X coordinate values of the steel rail at the positive direction of the X axis as a positive steel rail detection and calibration area, and counting all X coordinate values of the steel rail at the negative direction of the X axis as a negative steel rail detection and calibration area.

S33, calling standard detection information corresponding to the steel rail, determining Z-axis coordinates of all target points in the steel rail detection calibration area, comparing the Z-axis coordinates with the standard detection information to perform detection judgment, and determining an abnormal area in the steel rail detection calibration area.

The steel rail detection and calibration area is an area of the steel rail in a three-dimensional space actually. The Z-axis coordinates of all target points (all rail surfaces) in the rail detection calibration area (rail) are determined, and can be understood as the actual Z-axis coordinates of the rail surface.

It can be understood that the data of both rails are load standard and consistent, and standard detection information (standard coordinate values) corresponding to the rails is retrieved.

And comparing the actual Z-axis coordinate of the surface of the rail with the Z-axis coordinate in the standard detection information (standard coordinate value), thereby determining the abnormal area in the steel rail.

In some embodiments, the step S33 (retrieving the standard detection information corresponding to the steel rail, determining the Z-axis coordinates of all target points in the steel rail detection calibration area, comparing the Z-axis coordinates with the standard detection information to perform detection judgment, and determining the abnormal area in the steel rail detection calibration area) includes S331-S333:

and S331, comparing the Z-axis coordinate of each target point with the standard Z-axis coordinate in the standard detection information to obtain a Z-axis coordinate difference value, and if the Z-axis coordinate difference value is greater than the preset coordinate difference value, judging that the target point corresponding to the corresponding Z-axis coordinate is an abnormal target point.

The target point is an actual coordinate point at the steel rail. It will be appreciated that the Z-axis coordinate of the rail surface changes when large cracks or pits occur in the rail.

Therefore, the Z-axis coordinate difference is obtained by comparing the Z-axis coordinate at the steel rail with the standard Z-axis coordinate in the standard detection information, and it can be understood that a three-dimensional coordinate system is established, when cracks, depressions and the like appear on the surface of the steel rail in the forward direction of the Z-axis, the Z-axis coordinate at the target point is enlarged, and therefore the Z-axis coordinate difference after the difference is obtained is larger than 0; the preset coordinate difference may be an artificially preset coordinate difference according to the actual situation, and may be 0, etc., which are not limited herein.

It can be understood that, when the difference of the Z-axis coordinate is equal to 0, it indicates that the steel rail is not changed, and if the difference of the Z-axis coordinate is greater than 0, it indicates that the steel rail is changed, and it is an abnormal target point.

And S332, if the Z-axis coordinate difference is less than or equal to the preset coordinate difference, determining that the target point corresponding to the corresponding Z-axis coordinate is a normal target point.

It can be understood that if the difference value of the Z-axis coordinate is less than or equal to the preset coordinate difference value, the defect condition that the steel rail at the position has no depression, crack and the like for long-term use is indicated.

S333, calculating the distance between any two abnormal target points to obtain the distance between the target points, and classifying all the abnormal target points with the distance between the target points being smaller than the preset distance between the points into the same abnormal area.

It is understood that defects such as cracks, depressions, and the like are often abnormal regions composed of a plurality of abnormal target points.

Therefore, the distance between any two abnormal target points is calculated, when the distance between the two abnormal target points is smaller than the preset distance, the two abnormal target points are close to each other by default, and the two abnormal target points are classified into the same abnormal area. The preset distance may be a distance preset in advance by an operator, for example, 0.01, and is not limited herein.

In some embodiments, the steps S333 (calculating the distance between any two abnormal target points to obtain the distance between the target points, and classifying all the abnormal target points with the distance between the target points being smaller than the preset distance between the preset points into the same abnormal area) include steps S3331-S3334:

s3331, adding different area marks to all the abnormal target points, and calculating the distance between any two abnormal target points to obtain the distance between the target points.

The invention adds corresponding different area marks to all the abnormal target points to distinguish different abnormal target points conveniently, calculates the distance between any two points and judges whether the abnormal target points are in the same area conveniently according to the distance between the two points, such as whether the abnormal target points are in the same pit or crack.

For example: abnormal target point: A. b, C, D, E, etc. may also be 1, 2, 3, etc., and are not limited herein for distinguishing different abnormal target points and calculating the distance between target points between any two points.

S3332, synchronously modifying the area markers of all the abnormal target points with the distance between the target points smaller than the preset distance, wherein the synchronous modification is to copy the information of the area markers of other abnormal target points and add the information into the information of the corresponding area markers.

It can be understood that, when the distance between the target points is smaller than the preset distance, it indicates that the target points are close to each other and are at the same defect, and then the area markers of the two abnormal target points are modified synchronously in a manner that 2 copy the area markers of each other and put them into the information of the corresponding area markers again.

For example: and if the distance between the abnormal target point A and the abnormal target point B is smaller than the preset distance, copying the area mark B of the abnormal target point B by the abnormal target point A, copying the area mark A of the abnormal target point A by the abnormal target point B, wherein both the area marks are the area marks of AB, and replacing the area mark with the original A and B.

The server can directly judge whether the abnormal parts are in the same region through the region marks, and compared with the prior art, the method has the advantages that the graph defect is identified and processed, and then the classification is unified, so that the processing efficiency is better improved, and the method is simpler and quicker.

S3333, if the distance between the target point where the abnormal target point exists and the other abnormal target points is smaller than the preset distance, synchronously modifying the area markers of the corresponding abnormal target point and the other abnormal target points.

It is understood that the abnormal target point and other multiple abnormal target points may also be fused to form a complete abnormal area.

For example: if the distance between the abnormal target point C and one of the abnormal target points AB is smaller than the preset distance, the abnormal target point C and the abnormal target point AB are duplicated to obtain the area markers: ABC, so that the area marks of 3 dots are all ABC.

The server can directly classify and unify the region marks in a data processing mode, whether the region marks are in the same region or not is recognized in a region mark recognition mode, and the working efficiency is improved better

S3334, after determining that the area markers of all the abnormal target points are modified synchronously, classifying the abnormal target points with the same area marker into the same abnormal area.

It can be understood that after all the abnormal target points are synchronously modified, the server can directly classify the abnormal target points according to the area markers, and classify the abnormal target points into a class with the same area markers.

In other embodiments, the step S33 (retrieving standard detection information corresponding to the steel rail, determining Z-axis coordinates of all target points in the steel rail detection calibration area, comparing the Z-axis coordinates with the standard detection information to perform detection judgment, and determining an abnormal area in the steel rail detection calibration area) includes S334 to S337:

s334, counting 3 adjacent points of the X-axis coordinate and/or the Y-axis coordinate in the steel rail detection calibration area to obtain a plurality of inspection sets, and calculating the distance between the middle point and the points on the two sides in each inspection set in the Z axis to obtain a first distance value and a second distance value.

It can be understood that the rail can be divided into a rail having inclined surfaces on both sides and a rail having no inclined surface; the invention verifies the inclined plane in the form of 3 adjacent X-axis coordinates and/or Y-axis coordinates without considering the elimination of the inclined plane.

It should be noted that, in the invention, any 3 adjacent target points of the X-axis coordinate and/or the Y-axis coordinate in the rail region may be selected, and the verification inspection may be performed in the horizontal direction or the vertical direction, and the distance between the adjacent points on both sides and the middle point is calculated to obtain the first distance value and the second distance value.

And S335, extracting a first standard distance value and a second standard distance value in the standard detection information, and if the absolute values of the first distance value and the second distance value are respectively greater than the first standard distance value, and the absolute value of the difference between the first distance value and the second distance value is less than the second standard distance value, determining that the target point corresponding to the middle point in the inspection set is an abnormal target point.

It is understood that if the middle point and the points on both sides in the inspection set have a certain distance (fall) in the Z-axis direction, it is indicated that the depression may be an inclined surface here; that is, if the absolute values of the first distance value and the second distance value are respectively greater than the first standard distance value, the first distance value and the second distance value may be concave or inclined; the first standard distance value is a standard preset distance value, and the first standard distance value may be 0, which is not limited herein.

When the absolute value of the difference value between the first distance value and the second distance value is smaller than the second standard distance value, the area is determined to be a concave area; the second standard distance value may be 0.01, and is set to be a smaller allowable error value according to the actual situation.

As shown in fig. 2, for example: and checking three points in the sets A, B and C, wherein the coordinate value of the Z axis in the point A is 2, the coordinate value of the Z axis in the point C is 2, the coordinate value of the Z axis in the point B is 2.2, the difference between the point A and the point B is 2-2.2= -0.2, the difference between the point C and the point B is 2-2.2= -0.2, the absolute values of the two are both greater than 0, and the difference between the first distance value and the second distance value (-0.2) - (-0.2) =0 is smaller than the corresponding second standard distance value, so that the abnormal part such as a recess or a crack is determined.

S336, if the first distance value and the second distance value are respectively less than or equal to the first standard distance value, or the absolute value of the difference between the first distance value and the second distance value is greater than or equal to the second standard distance value, determining that the target point corresponding to the middle point in the inspection set is a normal target point.

It can be understood that, when the first distance value and the second distance value from the two side points to the middle point are respectively less than or equal to the first standard distance value, it indicates that the 3 points are all on the same horizontal plane; or the difference between the first distance value and the second distance value is greater than or equal to the second standard distance value, which indicates that 3 points are located at the inclined plane, and the target point corresponding to the middle point in the inspection set is determined to be a normal target point, that is, no corresponding defect exists.

For example: as shown in fig. 3A, three points in the set D, E, and F are checked, the coordinate value of the Z axis in the point D is 2, the coordinate value of the Z axis in the point E is 2, the coordinate value of the Z axis in the point F is 2, the difference between the point D and the point E is 0, and the difference between the point F and the point E is 0, so that all the 3 points in the set are on the same horizontal plane; as shown in fig. 3B, the three points in the set G, H, and I are checked, the coordinate value of the Z axis in the G point is 2.2, the coordinate value of the Z axis in the H point is 2.1, the coordinate value of the Z axis in the I point is 2, the difference between the I point and the H point is 2-2.1= -0.1, the difference between the G point and the H point is 2.2-2.1= -0.1, the absolute value of the difference between the two is 0.1- (-0.1) =0.2 greater than 0, and the target point corresponding to the middle point in the set is a normal target point.

S337, calculating the distance between any two abnormal target points to obtain the distance between the target points, and classifying all the abnormal target points with the distance between the target points smaller than the preset distance between the points into the same abnormal area.

It can be understood that, when determining the abnormal target point, the fusion processing needs to be performed on the abnormal target point to live in the corresponding abnormal region, the fusion mode principle is consistent, and the fusion is performed by using the region markers, which is not described herein again.

And S4, generating a corresponding processing display frame according to the attribute of the three-dimensional coordinates of the points in the abnormal area, combining the processing display frame and the abnormal area in the point cloud three-dimensional image, and generating a processing result.

The processing display frame is a to-be-processed display frame generated according to the abnormal area, and it can be understood that the server generates the corresponding processing display frame for highlighting according to the abnormal area, so that a user can directly observe the defective area conveniently.

Through the mode, the abnormal area is placed in the processing display frame and displayed in the point cloud three-dimensional image, and the convenience is brought to a user in the later period to directly observe and process.

In some embodiments, the step S4 (generating a corresponding processing display frame according to the property of the three-dimensional coordinates of the point in the abnormal region, displaying the processing display frame and the abnormal region in combination in the point cloud three-dimensional image, and generating a processing result) includes S41 to S44:

s41, determining the attribute of the three-dimensional coordinates of the points in each abnormal area, wherein the attribute of the three-dimensional coordinates of the points comprises the number of all the abnormal target points and the three-dimensional coordinates of each abnormal target point.

It can be understood that the server may determine the number of the abnormal target points and the three-dimensional coordinates of each abnormal target point in each abnormal area, so as to facilitate the subsequent generation of the corresponding processing display frame by using the number of the abnormal target points and the three-dimensional coordinates of each abnormal target point.

And S42, extracting an X-axis extreme value and a Y-axis extreme value in the three-dimensional coordinates of the abnormal target point corresponding to each abnormal area, and determining two groups of X-axis parallel lines and Y-axis parallel lines according to the X-axis extreme value and the Y-axis extreme value.

It can be understood that whether the defect area is a dent or a crack, the maximum value and the minimum value of the defect area on the plane can be obtained, and then corresponding parallel lines can be determined, so that the rectangular processing display frame can be conveniently generated.

S43, according to the frame line formed by the two groups of X-axis parallel lines and Y-axis parallel lines, the display color of the processing display frame is calculated according to the X-axis distance, the Y-axis distance and the number of the three-dimensional coordinates of all points, wherein the X-axis distance and the Y-axis distance are the X-axis distance and the Y-axis distance of the processing display frame.

It is understood that a rectangular region formed by intersecting parallel lines corresponding to the respective X-axis and Y-axis is generated as a processing display frame from the maximum value of the X-axis, the minimum value, and the maximum value and the minimum value of the Y-axis in the abnormal region.

It is understood that the X-axis distance can be determined by the difference between the X-axis extreme values (maximum and minimum values), and the Y-axis distance, that is, the side length of the processed display frame, can be determined by the difference between the Y-axis extreme values (maximum and minimum values).

Through the mode, a user can directly see the corresponding defect area through the display frame, and can judge the severity of the defect according to the different colors or the different color shades of the processed display frame.

In some embodiments, the step S43 (taking a frame line formed by the two sets of X-axis parallel lines and Y-axis parallel lines as a processing display frame, calculating the display color of the processing display frame according to the X-axis distance, the Y-axis distance, and the number of three-dimensional coordinates of all points, and according to the X-axis extreme value and the Y-axis extreme value, and the X-axis distance and the Y-axis distance of the processing display frame) includes S431 to S433:

and S431, determining a frame line formed by the intersection areas of the two groups of X-axis parallel lines and Y-axis parallel lines as a processing display frame, and amplifying the processing display frame by a preset multiple.

It should be noted that, because the cracks at the steel rails are different in size and smaller, the processing display frame is amplified by a corresponding preset multiple, wherein the preset multiple is the amplification multiple set in advance according to the actual situation; thereby facilitating the observation of the user and avoiding omission.

And S432, calculating according to the X-axis extreme value, the Y-axis extreme value and the preset multiple to obtain the X-axis distance and the Y-axis distance of the processing display frame after amplification, and obtaining the area of the processing display frame according to the X-axis distance and the Y-axis distance after amplification.

It can be understood that the X-axis distance can be determined by the difference between the X-axis extreme values (maximum and minimum), and the Y-axis distance can be determined by the difference between the Y-axis extreme values (maximum and minimum), that is, the side length of the display frame is processed, and then multiplied by the corresponding preset multiple, so as to obtain the size of the side length actually displayed, and thus the area of the display frame is processed according to the side length.

S433, obtaining three-dimensional coordinate density according to the area of the processing display frame and the number of the three-dimensional coordinates of the points, comparing the three-dimensional coordinate density with preset density intervals to obtain corresponding display colors, wherein each preset density interval has a corresponding display color, calculating the three-dimensional coordinate density through the following formula,

wherein ,

in the form of a three-dimensional coordinate density,

the number of the abnormal target points is,

is the maximum of the extreme values of the X-axis,

is the minimum of the extreme values of the X-axis,

the number of the first time is a preset multiple,

is the maximum of the extreme values of the Y-axis,

is the minimum of the Y-axis extrema,

to show the area of the frame, abnormalNumber of target points

And three-dimensional coordinate density

Proportional, three-dimensional coordinate density

And

inversely proportional, it is understood that the area of the display frame is processed

The larger and abnormal number of target points

The smaller the number, the smaller the crack, and the area of the display frame is treated in the opposite way

The smaller and abnormal target point number

The larger the number, the larger the crack, the density of the three-dimensional coordinate

The size of the value determines the color of the corresponding processing display box, for example: the larger the crack is, the darker the color of the corresponding processing display frame is, the smaller the crack is, the lighter the color of the corresponding processing display frame is, or different colors are set to represent different severity degrees, which is not limited herein.

And S44, determining a central point of the processing display frame and a central point of the abnormal region, and overlapping the central point of the processing display frame and the central point of the abnormal region to enable the processing display frame and the abnormal region to be combined and displayed in the point cloud three-dimensional image and generate a processing result, wherein the processing result comprises the number of the three-dimensional coordinates of the points.

It can be understood that the central point of the processing display frame and the central point of the abnormal region are simultaneously selected, and the central points of the processing display frame and the abnormal region are overlapped to enable the processing display frame and the abnormal region to be combined to be displayed in the point cloud three-dimensional image and display the number of corresponding abnormal target points.

The invention can not only display the defect area by frame selection and is convenient for a user to observe, but also display the number of corresponding abnormal target points, so that the user can check the defect area by combining numerical values with a processing display frame, and the size of the defect area can be more intuitively determined.

In order to better implement the three-dimensional point cloud calibration processing method for steel rail detection provided by the present invention, the present invention further provides a three-dimensional point cloud calibration processing apparatus for steel rail detection, as shown in fig. 4, including:

The present invention also provides a storage medium having a computer program stored therein, the computer program being executable by a processor to implement the methods provided by the various embodiments described above.

The storage medium may be a computer storage medium or a communication medium. Communication media includes any medium that facilitates transfer of a computer program from one place to another. Computer storage media may be any available media that can be accessed by a general purpose or special purpose computer. For example, a storage medium is coupled to the processor such that the processor can read information from, and write information to, the storage medium. Of course, the storage medium may also be integral to the processor. The processor and the storage medium may reside in an Application Specific Integrated Circuits (ASIC). Additionally, the ASIC may reside in user equipment. Of course, the processor and the storage medium may reside as discrete components in a communication device. The storage medium may be read-only memory (ROM), random-access memory (RAM), CD-ROMs, magnetic tapes, floppy disks, optical data storage devices, and the like.

The present invention also provides a program product comprising execution instructions stored in a storage medium. The at least one processor of the device may read the execution instructions from the storage medium, and the execution of the execution instructions by the at least one processor causes the device to implement the methods provided by the various embodiments described above.

In the above embodiments of the terminal or the server, it should be understood that the Processor may be a Central Processing Unit (CPU), other general-purpose processors, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of a method disclosed in connection with the present invention may be embodied directly in a hardware processor, or in a combination of hardware and software modules.

Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. The three-dimensional point cloud calibration processing method for steel rail detection is characterized by comprising the following steps:

2. The method for calibrating and processing the three-dimensional point cloud for steel rail detection according to claim 1,

the control point cloud collection equipment collects a point cloud three-dimensional image of the position where the steel rail is located at a preset height and a preset position, and determines point three-dimensional coordinates of all points in the point cloud three-dimensional image, and the control point cloud collection equipment comprises the following steps:

3. The method for calibrating and processing the three-dimensional point cloud for steel rail detection according to claim 2,

the method comprises the following steps of receiving size data configured by a user on a steel rail, wherein the size data at least comprise steel rail distance information and steel rail width information, calculating according to a three-dimensional coordinate of equipment, the steel rail distance information and the steel rail width information to obtain a target coordinate acquisition interval, and comprises the following steps:

4. The method for calibrating and processing the three-dimensional point cloud for steel rail detection according to claim 3,

determining a corresponding target area in the point cloud three-dimensional image as a steel rail detection calibration area according to the target coordinate acquisition interval, detecting and judging all point three-dimensional coordinates in the steel rail detection calibration area according to standard detection information, and determining an abnormal area in the steel rail detection calibration area, wherein the method comprises the following steps:

counting a target area formed by a target point corresponding to the positive X-axis coordinate as a positive steel rail detection calibration area, and counting a target area formed by a target point corresponding to the negative X-axis coordinate as a negative steel rail detection calibration area;

5. The method for calibrating and processing the three-dimensional point cloud for steel rail detection according to claim 4,

the method comprises the following steps of calling standard detection information corresponding to a steel rail, determining Z-axis coordinates of all target points in a steel rail detection calibration area, comparing the Z-axis coordinates with the standard detection information to perform detection judgment, and determining an abnormal area in the steel rail detection calibration area, wherein the method comprises the following steps:

comparing the Z-axis coordinate of each target point with the standard Z-axis coordinate in the standard detection information to obtain a Z-axis coordinate difference value, and if the Z-axis coordinate difference value is larger than a preset coordinate difference value, judging that the target point corresponding to the corresponding Z-axis coordinate is an abnormal target point;

6. The method for calibrating and processing the three-dimensional point cloud for detecting the steel rail according to claim 5, wherein before determining the corresponding target area in the point cloud three-dimensional image as the steel rail detection calibration area according to the target coordinate acquisition section, performing detection and judgment on three-dimensional coordinates of all points in the steel rail detection calibration area according to standard detection information, and determining an abnormal area in the steel rail detection calibration area, the method further comprises:

according to the preset width, coordinate point determination is respectively carried out in a corresponding negative direction and a positive direction by taking the positive termination X point coordinate and the negative termination X point coordinate as starting points, so that a second positive removal X point coordinate and a second negative removal X point coordinate are obtained;

7. The method for calibrating and processing the three-dimensional point cloud for steel rail detection according to claim 5,

the calculating the distance between any two abnormal target points to obtain the distance between the target points, and classifying all the abnormal target points with the distance between the target points being smaller than the preset distance between the points into the same abnormal area includes:

8. The method for calibrating and processing the three-dimensional point cloud for rail detection according to claim 7,

generating a corresponding processing display frame according to the attribute of the three-dimensional coordinates of the points in the abnormal area, combining the processing display frame with the abnormal area, displaying the processing display frame and the abnormal area in the point cloud three-dimensional image and generating a processing result, wherein the processing result comprises the following steps:

determining the attribute of a point three-dimensional coordinate in each abnormal area, wherein the attribute of the point three-dimensional coordinate comprises the number of all abnormal target points and the three-dimensional coordinate of each abnormal target point;

determining a central point of the processing display frame and a central point of the abnormal region, and setting the central point of the processing display frame to be overlapped with the central point of the abnormal region so as to combine the processing display frame and the abnormal region in the point cloud three-dimensional image to display and generate a processing result, wherein the processing result comprises the number of three-dimensional coordinates of points.

9. The method for calibrating and processing the three-dimensional point cloud for rail detection according to claim 8,

the frame line formed according to the two groups of X-axis parallel lines and Y-axis parallel lines is used as a processing display frame, and the display color of the processing display frame is calculated according to the X-axis distance, the Y-axis distance and the number of three-dimensional coordinates of all points according to the X-axis extreme value and the Y-axis extreme value and the X-axis distance and the Y-axis distance of the processing display frame, and the method comprises the following steps:

determining a frame line formed by the intersection areas of the two groups of X-axis parallel lines and Y-axis parallel lines as a processing display frame, and amplifying the processing display frame by a preset multiple;

wherein ,

in the form of a three-dimensional coordinate density,

the number of the abnormal target points is,

is the maximum of the extreme values of the X-axis,

is the minimum of the extreme values of the X-axis,

the number of the first time is a preset multiple,

is the maximum of the extreme values of the Y-axis,

the minimum value in the extreme values of the Y axis.

10. Three-dimensional point cloud calibration processing apparatus that rail detected, its characterized in that includes: