CN111307065A - Steel rail profile detection method, system and device based on polarization beam splitting - Google Patents

Steel rail profile detection method, system and device based on polarization beam splitting Download PDF

Info

Publication number
CN111307065A
CN111307065A CN202010147291.2A CN202010147291A CN111307065A CN 111307065 A CN111307065 A CN 111307065A CN 202010147291 A CN202010147291 A CN 202010147291A CN 111307065 A CN111307065 A CN 111307065A
Authority
CN
China
Prior art keywords
image
component
polarization
light
steel rail
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202010147291.2A
Other languages
Chinese (zh)
Other versions
CN111307065B (en
Inventor
方玥
王乐
王昊
王胜春
赵延峰
赵鑫欣
王宁
周谦
张翼
李海浪
黎国清
任盛伟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
China Academy of Railway Sciences Corp Ltd CARS
Infrastructure Inspection Institute of CARS
Beijing IMAP Technology Co Ltd
Original Assignee
China Academy of Railway Sciences Corp Ltd CARS
Infrastructure Inspection Institute of CARS
Beijing IMAP Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by China Academy of Railway Sciences Corp Ltd CARS, Infrastructure Inspection Institute of CARS, Beijing IMAP Technology Co Ltd filed Critical China Academy of Railway Sciences Corp Ltd CARS
Priority to CN202010147291.2A priority Critical patent/CN111307065B/en
Publication of CN111307065A publication Critical patent/CN111307065A/en
Application granted granted Critical
Publication of CN111307065B publication Critical patent/CN111307065B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B11/00—Measuring arrangements characterised by the use of optical techniques
    • G01B11/24—Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures
    • G01B11/25—Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures by projecting a pattern, e.g. one or more lines, moiré fringes on the object
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/13—Edge detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G06T7/66—Analysis of geometric attributes of image moments or centre of gravity
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20212—Image combination
    • G06T2207/20221—Image fusion; Image merging
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30108—Industrial image inspection

Abstract

The invention discloses a method, a system and a device for detecting steel rail profile based on polarization beam splitting, wherein the method comprises the following steps: the method comprises the steps of obtaining a P component polarization image and an S component polarization image of steel rail surface reflected light, wherein the steel rail surface reflected light is decomposed into P component transmission light and S component reflected light which are perpendicular to each other in a vibration plane through a polarization beam splitter prism, the P component polarization image is an image formed by the P component transmission light, and the S component polarization image is an image formed by the S component reflected light; determining a total intensity image of the reflected light on the surface of the steel rail according to the P component polarization image and the S component polarization image; performing fusion processing on the P component polarization image, the S component polarization image and the total intensity image to obtain a fusion image; and extracting the steel rail outline data according to the fused image. The method can obtain the high-quality steel profile light bar image without local overexposure, extract the accurate light bar center from the high-quality steel profile light bar image, and improve the accuracy of steel profile detection.

Description

Steel rail profile detection method, system and device based on polarization beam splitting
Technical Field
The invention relates to the field of steel rail detection, in particular to a method, a system and a device for detecting steel rail profile based on polarization beam splitting.
Background
This section is intended to provide a background or context to the embodiments of the invention that are recited in the claims. The description herein is not admitted to be prior art by inclusion in this section.
Because the change of the steel rail profile is directly related to the safe operation of the railway track, the detection of the steel rail profile is beneficial to mastering the service state of the steel rail and guiding the grinding operation of the steel rail, and is an important means for railway operation and maintenance.
The principle of the steel rail profile detection is that actually detected steel rail profile data is compared with standard steel rail profile data, so that parameters such as vertical abrasion loss, side surface abrasion loss and the like of the steel rail are obtained. At present, there are two main types of rail profile detection means: one is contact detection, in which a probe is contacted with a steel rail to detect the profile data of the steel rail; the other type is non-contact detection, and the intensity information of the reflected light on the surface of the steel rail is used for extracting the profile data of the steel rail.
For contact detection, the defects of low detection efficiency, high labor cost and the like exist because the probe is required to be contacted with the steel rail. For non-contact detection, due to the fact that the working condition of a railway site is poor, after a steel rail is in service for a period of time, the surface state of the steel rail can be changed (for example, the surface is not smooth, foreign matters on the surface, a steel rail light band, a rail head are rusted and the like), and therefore energy distribution of reflected light on the surface of the steel rail is interfered, and energy distribution is abnormal. Particularly, after the steel rail is polished, the problem of excessive exposure of a local area (namely, the problem of overexposure) exists in the acquired light bar image of the profile of the steel rail, so that the centers of the light bars are difficult to accurately extract, and the error of the detection result of the profile of the steel rail is large.
Fig. 1 is a light bar image formed by collecting light reflected from the surface of a polished steel rail by using a conventional steel rail profile detection method, and it can be seen that, for the polished steel rail, the surface appearance and roughness of the polished steel rail are changed, which affects the distribution of specular reflection energy and diffuse reflection energy on the surface of the steel rail, when the specular reflection light direction of a local area is consistent with or close to the optical axis of an imaging lens, a large amount of specular reflection light enters an image detector, so that an overexposure phenomenon occurs in the light bar image of the profile of the steel rail collected by the image detector, for example, in an overexposure area shown by a dotted line rectangular frame in fig. 1. In an overexposure area, due to the fact that the energy distribution of the light bars is abnormal, the extraction error of the centers of the light bars is large, the accurate centers of the light bars cannot be obtained frequently, and therefore the detection result of the profile of the steel rail has large errors.
Fig. 2 shows the result of extracting the light bar center from the light bar image shown in fig. 1 by using the maximum value method, the gray scale center-of-gravity method, and the Steger method, respectively. As shown in fig. 2, an icon a shows a center of a light stripe extracted by a maximum value method, an icon b shows a center of a light stripe extracted by a gray scale gravity center method, an icon c shows a center of a light stripe extracted by a Steger method, and an icon d shows a result of mapping the profile data of the steel rail measured by a minipro profiler into image coordinates. The miniprox rail profile gauge can realize contact measurement of the rail profile and the measurement result is accurate, so that the miniprox rail profile gauge can be used as a reference standard for different light bar center extraction methods.
As can be seen from fig. 2, in the non-overexposure area (i.e., the area outside the dashed rectangular frame), the profile of the steel rail obtained by using the different light bar center extraction methods is substantially the same as the profile of the steel rail measured by the miniprox profile gauge; in an overexposure area (namely an area in a dotted line rectangular frame), due to the fact that energy distribution of cross sections of light bars is complex, errors of different degrees occur in the centers of the light bars obtained by adopting different light bar center extraction methods, and the errors of different degrees also occur in a steel rail profile measurement result, wherein the maximum error of a gray scale gravity center method reaches 0.95mm, and even a Steger method with the highest precision also generates an error of 0.29 mm.
Therefore, the conventional steel rail profile detection method based on line structured light directly collects the intensity information of reflected light on the surface of a steel rail when the steel rail is in a severe working condition, particularly a newly polished steel rail, and a formed light strip image has a local overexposure phenomenon. In an overexposure area, the imaging quality of the optical strips is poor, and the energy distribution of the optical strips is complex, so that the real steel rail profile information is interfered, a stable and accurate optical strip center cannot be obtained, and finally, the error of the steel rail profile detection result is large.
Although the problem of local overexposure can be solved by reducing the exposure time, insufficient exposure of the light bars in the normal exposure area in the same image can be caused at the same time, and finally the extraction of the light bars in the normal area is influenced, so that the problem of local overexposure of the image of the light bars with the steel rail profile cannot be effectively solved only by reducing the exposure time.
In view of the above problems, no effective solution has been proposed.
Disclosure of Invention
The embodiment of the invention provides a method for detecting a steel rail profile based on polarization beam splitting, which is used for solving the technical problems that a light bar image acquired by the existing method for detecting the steel rail profile has a local overexposure phenomenon, the center of the light bar is difficult to accurately extract, and the error of a steel rail profile detection result is larger, and comprises the following steps: the method comprises the steps of obtaining a P component polarization image and an S component polarization image of steel rail surface reflected light, wherein the steel rail surface reflected light is decomposed into P component transmission light and S component reflected light which are perpendicular to each other in a vibration plane through a polarization beam splitter prism, the P component polarization image is an image formed by the P component transmission light, and the S component polarization image is an image formed by the S component reflected light; determining a total intensity image of the reflected light on the surface of the steel rail according to the P component polarization image and the S component polarization image; performing fusion processing on the P component polarization image, the S component polarization image and the total intensity image to obtain a fusion image; and extracting the steel rail outline data according to the fused image.
The embodiment of the invention also provides a steel rail profile detection system based on polarization beam splitting, which is used for solving the technical problems that the light bar image acquired by the existing steel rail profile detection method has local overexposure and overexposure phenomena, the centers of the light bars are difficult to accurately extract, and the error of the steel rail profile detection result is larger, and the system comprises: rail profile shape determine module and image processing host computer, rail profile shape determine module includes: the device comprises a line structured light laser, a polarization splitting prism, a first imaging device and a second imaging device; the linear structure light laser is used for emitting light to the surface of the steel rail, and the light cutter plane is vertically incident to the surface of the steel rail; the polarization beam splitter prism is used for collecting the surface reflected light of the steel rail and splitting the surface reflected light of the steel rail into P component transmitted light and S component reflected light with mutually vertical vibration planes; the first imaging device is used for acquiring the P component transmission light and generating a P component polarization image; the second imaging device is used for acquiring S-component reflected light and generating an S-component polarized image; and the image processing host is communicated with the first imaging device and the second imaging device respectively, and is used for determining a total intensity image of the reflected light on the surface of the steel rail according to the P component polarization image generated by the first imaging device and the S component polarization image generated by the second imaging device, fusing the P component polarization image, the S component polarization image and the total intensity image to obtain a fused image, and extracting the profile data of the steel rail according to the fused image.
The embodiment of the invention also provides a steel rail profile detection device based on polarization beam splitting, which is used for solving the technical problems that the light bar image acquired by the existing steel rail profile detection method has local overexposure phenomenon, the light bar center is difficult to accurately extract, and the error of the steel rail profile detection result is larger, and the device comprises: the device comprises an image acquisition unit, a polarization splitting prism and a polarization splitting prism, wherein the image acquisition unit is used for acquiring a P component polarization image and an S component polarization image of the reflected light on the surface of the steel rail, the reflected light on the surface of the steel rail is decomposed into a P component transmission light and an S component reflected light which are mutually vertical to a vibration plane through the polarization splitting prism, the P component polarization image is an image formed by the P component transmission light, and the S component polarization image is an image formed by the S component reflected; the first image processing unit is used for determining a total intensity image of the reflected light on the surface of the steel rail according to the P component polarization image and the S component polarization image; the second image processing unit is used for carrying out fusion processing on the P component polarization image, the S component polarization image and the total intensity image to obtain a fusion image; and the steel rail profile extraction unit is used for extracting the steel rail profile data according to the fusion image.
The embodiment of the invention also provides computer equipment for solving the technical problem that the error of the detection result of the steel rail profile is larger due to the fact that the light bar image acquired by the conventional steel rail profile detection method has a local overexposure phenomenon and the center of the light bar is difficult to accurately extract.
The embodiment of the invention also provides a computer readable storage medium, which is used for solving the technical problem that the error of the steel rail profile detection result is larger due to the fact that the light bar image acquired by the existing steel rail profile detection method has a local overexposure phenomenon and the center of the light bar is difficult to accurately extract.
In the embodiment of the invention, the steel rail surface reflected light is decomposed into the P component transmitted light and the S component reflected light with mutually vertical vibration planes by using the polarization beam splitter prism, so that a P component polarization image formed by the P component transmitted light and an S component polarization image formed by the S component reflected light are obtained, the total intensity image of the steel rail surface reflected light is determined according to the P component polarization image and the S component polarization image, then the P component polarization image, the S component polarization image and the total intensity image are subjected to fusion processing, and finally, the steel rail profile data are extracted according to the fusion image.
According to the embodiment of the invention, the P component polarization image, the S component polarization image and the total intensity image of the steel rail profile light bar are fused, so that a high-quality steel rail profile light bar image without a local overexposure phenomenon can be obtained, an accurate light bar center is extracted from the high-quality steel rail profile light bar image, and the accuracy of steel rail profile detection is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts. In the drawings:
FIG. 1 is a schematic view of a light bar image formed by collecting light reflected from a ground surface of a rail according to the prior art;
fig. 2 is a schematic diagram illustrating a result of extracting light strip centers of a polished light strip image of a rail by using a plurality of light strip center extraction algorithms according to the prior art;
fig. 3 is a schematic diagram of a principle of detecting a rail profile based on polarization beam splitting according to an embodiment of the present invention;
fig. 4 is a schematic diagram of a rail profile detection assembly based on polarization splitting according to an embodiment of the present invention;
fig. 5 is a schematic diagram of a rail profile detection system based on polarization beam splitting according to an embodiment of the present invention;
fig. 6 is a flowchart of a method for detecting a profile of a steel rail based on polarization beam splitting according to an embodiment of the present invention;
fig. 7 is a flowchart of an alternative rail profile detection method based on polarization beam splitting according to an embodiment of the present invention;
FIG. 8 is a schematic diagram of an image fusion algorithm provided in an embodiment of the present invention;
FIG. 9 is a schematic diagram of a P-component polarization image provided in an embodiment of the present invention;
FIG. 10 is a schematic diagram of an S-component polarization image provided in an embodiment of the present invention;
FIG. 11 is a schematic representation of a total intensity image provided in an embodiment of the present invention;
FIG. 12 is a schematic diagram of a fused image provided in an embodiment of the present invention;
FIG. 13 is a diagram illustrating a result of extracting light bar centers from a fused image according to an embodiment of the present invention;
FIG. 14 is a schematic diagram illustrating an error of a rail profile detection result according to an embodiment of the present invention;
fig. 15 is a schematic view of a rail profile detection apparatus based on polarization splitting according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the embodiments of the present invention are further described in detail below with reference to the accompanying drawings. The exemplary embodiments and descriptions of the present invention are provided to explain the present invention, but not to limit the present invention.
In the description of the present specification, the terms "comprising," "including," "having," "containing," and the like are used in an open-ended fashion, i.e., to mean including, but not limited to. Reference to the description of the terms "one embodiment," "a particular embodiment," "some embodiments," "for example," etc., means that a particular feature, structure, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the application. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. The sequence of steps involved in the embodiments is for illustrative purposes to illustrate the implementation of the present application, and the sequence of steps is not limited and can be adjusted as needed.
As can be seen from the description of the background of the present invention, in the conventional method for detecting the profile of a rail based on line structured light, the light bar image formed according to the intensity information of the reflected light on the surface of the rail is prone to have a local overexposure phenomenon, and this phenomenon is particularly serious for the detection of the polished rail. Due to the fact that the light bar image subjected to local overexposure cannot be extracted to the accurate light bar center, the error of the detection result of the steel rail profile is large.
In order to solve the above problems, embodiments of the present invention provide a rail profile detection scheme based on polarization splitting, which utilizes a polarization splitting prism to split a reflected light on a rail surface into two polarized lights (i.e., a P-component transmitted light and an S-component reflected light) with mutually perpendicular vibration directions, and the two polarized lights are received by two imaging devices to form a P-component polarized image and an S-component polarized image of a rail profile light bar, while a total intensity image of the rail profile light bar can be calculated from the P-component polarized image and the S-component polarized image, and then a fused image with high image quality is obtained by fusing the P-component polarized image, the S-component polarized image, and the total intensity image of the rail profile light bar (using a region of a normal light bar in the polarized image to replace a light bar overexposed region in the total intensity image).
Because the quality of the steel rail profile light stripe in the fused image is obviously improved, the noise interference of an overexposure area is eliminated, the local overexposure phenomenon does not occur in the steel rail profile light stripe image any more, and the real profile of the steel rail can be more accurately reflected by the light stripe center extraction result. The high-quality steel rail profile light strip image not only reduces the difficulty of later-stage light strip center extraction, but also improves the accuracy of steel rail profile detection under severe working conditions.
Fig. 3 is a schematic diagram of a principle of detecting a rail profile based on polarization splitting according to an embodiment of the present invention, as shown in fig. 3, a light knife plane emitted by a linear structured light laser 1 is perpendicularly incident on a surface of a rail 6, and a reflected light thereof passes through a polarization splitting prism 4 and is then decomposed into a P component and an S component perpendicular to each other by the polarization splitting prism 4, where the P component is a transmitted light, the vibration plane is in a paper plane, and the S component is a mirror reflected light and vibrates in a plane perpendicular to the paper plane. P component transmission light is incident to a first area array camera 2-1 after passing through a first imaging lens 3-1, S component is incident to a second area array camera 2-2 after passing through a second imaging lens 3-2, and P component polarized image and S component polarized image of steel rail profile light bar are obtained by the first area array camera 2-1 and the second area array camera 2-2 and are respectively marked as IpAnd Is. Obtaining a total intensity image from the P component polarization image and the S component polarization image, and recording the total intensity image as ItIt can be calculated by formula (1):
It＝Ip+Is(1)
it should be noted that the total intensity image obtained by the above formula (1) is a light bar image formed by collecting light reflected from the surface of the steel rail by using a conventional steel rail profile detection method. Under complex working conditions, the problem of local overexposure easily occurs in the total intensity image, the extraction precision of the light strip center is influenced by the overexposure area, and finally the accuracy of the detection of the steel rail profile is reduced. The P component polarization image and the S component polarization image do not have the problem of local overexposure because the polarization information of the laser reflected by the surface of the steel rail is acquired, but the contrast ratio is reduced compared with the total intensity image. Therefore, 3 images are fused by an image fusion algorithm through reducing the weight of the overexposure area and improving the weight of the normal area, the area with the overexposure light bars in the intensity image is replaced by the area with the normal light bars in the polarization image, a fused image with high image quality is obtained, the area with the overexposure light bars in the fused image does not exist any more, and the problem of local overexposure in the traditional method is solved.
Optionally, as shown in fig. 4, a set of steel rail profile detection assembly 5 based on the polarization splitting principle may be formed by main components such as the line structured light laser 1, the polarization splitting prism 4, the first imaging lens 3-1, the second imaging lens 3-2, the first area array camera 2-1, and the second area array camera 2-2, wherein an included angle between an optical axis of the first imaging lens 3-1 and a symmetric center line of the line structured light laser 1 is α, a value range is 20 ° to 60 °, and optical axes of the first imaging lens 3-1 and the second imaging lens 3-2 are perpendicular to each other.
Further, as shown in fig. 5, a rail profile detection system can be formed by main components such as a rail profile detection assembly 5, a detection beam 7, a signal synchronization unit 9, a photoelectric encoder 10, and an image processing host 8, and the system includes four sets of rail profile detection assemblies 5, the four sets of rail profile detection assemblies 5 are rigidly connected through the detection beam 7 and are respectively located on the left and right sides of two rails (the icon 6-1 is a left strand of rail, and the icon 6-2 is a right strand of rail), wherein a laser plane in the rail profile detection assembly 5 is perpendicular to a rail direction, an included angle between a symmetric center line of the line structured light laser 1 and a rail vertical direction is β is in a range of 30-50 degrees, the photoelectric encoder 10 is used for profile equidistant sampling, and the signal synchronization unit 9 sends out a synchronization signal to control eight area array cameras in the four sets of profile detection assemblies to simultaneously acquire rail profile images.
Based on the above inventive concept, an embodiment of the present invention provides a method for detecting a rail profile based on polarization splitting, and fig. 6 is a flowchart of the method for detecting a rail profile based on polarization splitting, as shown in fig. 6, the method may include the following steps:
s601, obtaining a P component polarization image and an S component polarization image of the steel rail surface reflected light, wherein the steel rail surface reflected light is decomposed into P component transmission light and S component reflected light which are perpendicular to each other in a vibration plane through a polarization beam splitter prism, the P component polarization image is an image formed by the P component transmission light, and the S component polarization image is an image formed by the S component reflected light.
It should be noted that, after the reflected light on the surface of the steel rail passes through the polarization beam splitter prism, the reflected light is decomposed into two polarized lights (i.e., P-component transmitted light and S-component reflected light) whose vibration planes are perpendicular to each other, and therefore, the two polarized lights are simultaneously collected by the two imaging devices, and two polarized images (i.e., P-component polarized image and S-component polarized image) can be obtained. Because the polarization image does not have the local overexposure phenomenon, but the imaging contrast of the polarization image is lower than that of the total intensity image, in order to obtain the steel rail profile light bar image with high imaging contrast and no local overexposure phenomenon, the normal light bar area in the polarization image can be used for replacing the light bar area with the overexposure phenomenon in the total intensity image.
And S602, determining a total intensity image of the reflected light on the surface of the steel rail according to the P component polarization image and the S component polarization image.
It should be noted that, in the embodiment of the present invention, the P-component polarization image and the S-component polarization image are images acquired from the same rail profile cross section at the same time, and have the characteristic of pixel-level alignment, so that the total intensity image of the reflected light on the rail surface can be calculated by the above formula (1).
And S603, carrying out fusion processing on the P component polarization image, the S component polarization image and the total intensity image to obtain a fusion image.
It should be noted that, because the characteristic information of the optical strips in the P-component polarization image, the S-component polarization image and the total intensity image has a difference, the P-component polarization image and the S-component polarization image do not have the problem of local overexposure because the polarization information of the laser reflected by the surface of the steel rail is acquired, but the contrast ratio is reduced compared with the total intensity image, so that the P-component polarization image, the S-component polarization image and the total intensity image are fused by reducing the weight of the overexposure region and increasing the weight of the normal region, and an image fusion algorithm is constructed to fuse the P-component polarization image, the S-component polarization image and the total intensity image, so that the overexposure region of the optical strips in the total intensity image can be replaced by the normal region of the optical strips in the polarization image, a.
As an optional implementation manner, when the P-component polarization image, the S-component polarization image, and the total intensity image are subjected to the fusion processing, the fusion weight of each image to be fused may be determined according to the feature information of the light bars in each image to be fused. Thus, in an alternative embodiment, the step S603 may be implemented by: determining the fusion weight of the P component polarization image, the S component polarization image and the total intensity image according to the characteristic information of the light bars in the P component polarization image, the S component polarization image and the total intensity image; and according to the fusion weight of the P component polarization image, the S component polarization image and the total intensity image, carrying out fusion processing on the P component polarization image, the S component polarization image and the total intensity image to obtain a fusion image.
When the fusion weight of the P component polarization image, the S component polarization image and the total intensity image is determined, the fusion weight of the polarization images of different light bars can be determined by any one of the following modes, ① determining the fusion weight of the polarization images of different light bars according to the width information of the light bars in the polarization images of different light bars, ② determining the fusion weight of the polarization images of different light bars according to the intensity information of the light bars in the polarization images of different light bars.
Alternatively, the above S603 may perform the fusion process on the P component polarization image, the S component polarization image, and the total intensity image by formula (2):
Figure BDA0002401202840000081
wherein, FkjRepresenting the gray value of the jth column and kth row pixels in the fusion image F;
Figure BDA0002401202840000082
representing the gray value of the jth column and kth row of pixels in the ith image to be fused;
Figure BDA0002401202840000083
representing the fusion weight of the jth pixel in the ith image to be fused; n represents the number of images to be fused.
Further, the fusion weight of the different light bar polarization images can be determined by equation (3):
Figure BDA0002401202840000091
wherein the content of the first and second substances,
Figure BDA0002401202840000092
representing the width of a light bar in the jth column in the ith image to be fused; mrRepresenting a reference width of the light bar;
Figure BDA0002401202840000093
representing the difference value between the light strip width of the jth column in the ith image to be fused and the reference width of the light strip; delta0Is shown as
Figure BDA0002401202840000094
A number set when equal to 0 to exclude the denominator being 0; m represents a weight adjustment factor.
And S604, extracting the steel rail outline data according to the fusion image.
It should be noted that the fused image for extracting the rail profile data in S604 is obtained by fusing the total intensity image and the polarization image of the rail profile light bar, so that the imaging contrast can be ensured and the overexposure phenomenon can be avoided. Since the rail profile data is determined according to the light bar center extracted from the light bar image, S604 may be implemented by: extracting the position information of the center of the optical strip in the fusion image; coordinate conversion is carried out on the position information of the center of the optical strip in the fusion image, and the profile data of the half section of one side of the steel rail is obtained; and carrying out profile splicing treatment on the profile data of the half sections on the two sides of the steel rail to obtain the profile data of the full section of the steel rail.
Optionally, when extracting the position information of the center of the light bar in the fused image, the position information of the center of the light bar in the fused image may be extracted by using any one of the following algorithms: the maximum method, the gray scale center of gravity method, and the Steger method.
As can be seen from the above, in the rail profile detection method based on polarization splitting provided in the embodiment of the present invention, the polarization splitting prism is used to split the reflected light on the surface of the rail into the P-component transmitted light and the S-component reflected light with the vibration planes perpendicular to each other, so as to obtain the P-component polarization image formed by the P-component transmitted light and the S-component polarization image formed by the S-component reflected light, determine the total intensity image of the reflected light on the surface of the rail according to the P-component polarization image and the S-component polarization image, perform fusion processing on the P-component polarization image, the S-component polarization image, and the total intensity image, and finally extract the rail profile data according to the fusion image.
According to the method for detecting the steel rail profile based on the polarization beam splitting, provided by the embodiment of the invention, the P component polarization image, the S component polarization image and the total intensity image of the steel rail profile light bar are fused, so that a high-quality steel rail profile light bar image without a local overexposure phenomenon can be obtained, an accurate light bar center is extracted from the high-quality steel rail profile light bar image, and the accuracy of the detection of the steel rail profile is improved.
Fig. 7 is a flowchart of an alternative rail profile detection method based on polarization splitting according to an embodiment of the present invention, as shown in fig. 7, including the following steps:
and S701, acquiring a P component polarization image, an S component polarization image and a total intensity image.
Specifically, a triggering source is composed of a photoelectric encoder and a signal synchronization unit, four sets of profile detection assemblies are synchronously triggered to sample at equal intervals, two area-array cameras in each set of profile detection assemblies are used for simultaneously collecting the light bar images of the steel rail half-section profile, two polarization images with mutually perpendicular vibration directions, namely a P component polarization image and an S component polarization image, are obtained, and the total intensity image of the steel rail profile light bar is calculated by the P component polarization image and the S component polarization image.
S702, evaluating the fusion weight of each image to be fused based on the light bar reliability.
Specifically, the fusion weight of the P-component polarization image, the S-component polarization image and the total intensity image is calculated on the basis of reducing the fusion weight of the overexposed region and increasing the fusion weight of the normal region.
And S703, constructing an image fusion algorithm to obtain a fusion image.
Specifically, according to the fusion weight of the P component polarization image, the S component polarization image and the total intensity image, the P component polarization image, the S component polarization image and the total intensity image are fused, so that the problem of local overexposure does not occur in the fused image.
And S704, extracting the centers of the light bars.
Specifically, for the fused rail profile light bar image, accurate light bar center position information is obtained by using a traditional light bar center extraction method, such as a gray scale gravity center method, a Steger method, template matching and the like.
And S705, calculating the profile of the full section of the steel rail.
Specifically, the half-section profile data of the steel rail is inversely calculated according to the central position of the steel rail profile light bar and the internal and external parameters of a camera calibrated in advance. For the single-stranded steel rail, the left and right profile detection assemblies are used for simultaneously acquiring the profile information of the left and right half sections of the steel rail, and profile splicing is carried out based on the position relation between the left and right half section profile data and the two side profile detection assemblies to obtain the relatively complete profile data of the single-stranded steel rail.
It should be noted that, in order to evaluate the light strip quality of each image to be fused and determine the fusion weight of each image to be fused, the embodiment of the present invention introduces a light strip reliability evaluation mechanism. The light bar reliability reflects the quality of the light bar image of the steel rail profile and is used for evaluating the reliability of the central position of the light bar. Under normal working conditions, the width of the light bar in the steel rail profile image is in a certain range, and the width of the light bar in the overexposure area is obviously increased, so that the credibility of the light bar can be evaluated according to the width of the light bar in the steel rail profile image. In addition to the light stripe width, the total intensity within the light stripe width reflects the quality of the light stripe, and the total intensity within the light stripe width is within a certain range for a normal light stripe image, while the total intensity is significantly increased for an overexposed region due to overexposure, and thus the total intensity within the light stripe width can also be used as a characteristic amount for evaluating the reliability of the light stripe. Therefore, the embodiment of the invention selects the width of the light strip or the total intensity in the width of the light strip as the characteristic quantity of the evaluation of the light strip reliability, evaluates the light strip reliability of each image and further calculates the image fusion weight. Alternatively, the two feature quantities may be combined to calculate the image fusion weight. The embodiment of the invention adopts a mode of fusing the images in rows to calculate the light bar reliability of each row of each image.
The following explains the fusion process of three images, i.e., the P-component polarization image, the S-component polarization image and the total intensity image, with the light stripe width as an example of light stripe reliability evaluation, and as shown in fig. 8, for the jth row of the ith light stripe polarization image, the light stripe width of the row is calculated and recorded as
Figure BDA0002401202840000111
Then, the width of the row of light bars and the reference width M of the light bars are calculatedrDifference of (2)
Figure BDA0002401202840000112
Difference value
Figure BDA0002401202840000113
The smaller the light bar is, the closer the light bar is to the reference light bar, the higher the imaging quality and the light bar reliability, and the higher the fusion weight should be when the images are fused, so the above formula (3) is adopted to calculate the fusion weight of the current column of the image
Figure BDA0002401202840000115
When in use
Figure BDA0002401202840000114
Is offset by a particularly small number Δ when equal to 00To exclude the case where the denominator is 0, Δ0The value range is 0.001-0.1.
Where i is 1,2,3, and 4 respectively denote light bar polarization images in four directions, and j is 1,2,3, …, Width, and Width denotes the number of columns of the image to be fused. M represents a weight adjustment factor, the value range of M is 1-3, the larger the value of M is, the higher the weight of a column which is closer to a reference light bar is, and the reference width M of the light barrThe width of the light strip under normal working conditions can be determined, and the width of the light strip during calibration can also be determined.
For each row of pixels of the jth column of the ith image, the gray value of each row is used
Figure BDA0002401202840000116
Multiplying by the fusion weight of the column
Figure BDA0002401202840000117
Summing to obtain the gray value F of the kth row pixel of the jth column in the fusion image FkjThe fusion image F is obtained as shown in formula (2), where k is 1,2,3, …, Height, and Height represents the Height of the image to be fused.
In the following, theoretical analysis is performed on the better detection effect of the steel rail profile that can be obtained by the method for detecting the steel rail profile provided by the embodiment of the invention:
the traditional steel rail profile detection technology acquires the intensity information of the light bar image of the steel rail profile, the steel rail profile is obtained by utilizing the light bar intensity information, when the surface state of the steel rail is good, the intensity distribution of the light bar image of the profile is reasonable, the section intensity basically accords with Gaussian distribution, the extraction result of the light bar is accurate, and the profile detection result is closer to the real profile of the steel rail. However, when the railway site encounters a complex working condition, such as the existence of factors such as unevenness of the surface of the steel rail, surface foreign matters, grinding of the steel rail and the like, the energy distribution of laser reflected by the surface of the steel rail is disturbed, so that an overexposure phenomenon easily occurs on an optical strip image, the cross-sectional intensity distribution of an overexposure area is disordered and has large noise interference, the profile of the steel rail cannot be well reflected, and the profile detection error is large.
Compared with the traditional intensity imaging technology, the polarization imaging technology can not only obtain the intensity information of a measured object, but also obtain the polarization information of more dimensions of the measured object, integrates the polarization information and the intensity information, can reduce the influence of specular reflection light, and improves the imaging contrast ratio, so the embodiment of the invention is based on the traditional line structured light steel rail profile detection technology, divides the laser reflected by the surface of a steel rail into two beams of polarized light with mutually vertical vibration directions through a polarization beam splitter prism, simultaneously receives the two beams of polarized light by two common planar array cameras to obtain a P component polarization image and an S component polarization image of a steel rail profile light bar, obtains a total intensity image from the P component polarization image and the S component polarization image, and further obtains a fused steel rail profile light bar image by fusing the P component polarization image, the S component polarization image and the total intensity image of the steel rail profile light bar, the method can solve the problem that local area exposure is excessive in the light bar image obtained by the traditional steel rail profile detection method, and is particularly suitable for the situation of steel rail profile detection after polishing.
The P-component polarization image shown in fig. 9 and the S-component polarization image shown in fig. 10 can be obtained by photographing the same position of the rail, and the total intensity image shown in fig. 11 can be calculated from the P-component polarization image and the S-component polarization image. Fig. 12 is a fused image obtained by fusing the P-component polarization image, the S-component polarization image, and the total intensity image according to the embodiment of the present invention. As can be seen from fig. 12, the fused light bar image of the rail profile has no overexposed area, and the imaging quality is significantly improved.
Fig. 13 shows the result of extracting the center of light in the fused image, and belongs to the same position as the overexposed region in fig. 2. As shown in fig. 13, an icon a is a result of extracting light bars by using a maximum value method, an icon b is a result of extracting light bars by using a gray scale gravity center method, an icon c is a result of extracting light bars by using a Steger method, and an icon d is a result of mapping measurement data of a minipro profiler into an image coordinate system.
As can be seen by comparing fig. 13 and fig. 2, compared with the light bar center extracted from the light bar image by the conventional steel rail profile detection method, the embodiment of the invention has no local overexposure phenomenon in the fused image, has higher accuracy of the light bar center position, has the best consistency with the minipro profile instrument measurement, and can more accurately reflect the real profile of the steel rail. In order to quantitatively analyze the advantages of the steel rail profile detection method in the embodiment of the invention, for the same position of the light bar image obtained by the conventional steel rail profile detection method where the light bar is overexposed, the measurement result of the minipro profiler is used as a reference value, and the measurement errors of the steel rail profile obtained by the steel rail profile detection method and the conventional steel rail profile detection method provided by the embodiment of the invention are respectively calculated, the results are shown in fig. 14, a-1, a-2 and a-3 in fig. 14 respectively represent the measurement deviation of the steel rail profile obtained by the conventional steel rail detection method adopting a maximum light bar center extraction algorithm, a gray scale gravity center maximum light bar center extraction algorithm and a Steger maximum light bar center extraction algorithm, and B-1, B-2 and B-3 in fig. 14 respectively represent the steel rail detection method provided by the embodiment of the invention respectively adopting a maximum light bar center extraction algorithm, a local light bar center extraction, And the steel rail profile measurement deviation is obtained by a gray gravity center maximum light bar center extraction algorithm and a Steger maximum light bar center extraction algorithm.
As can be seen from fig. 14, compared with the conventional rail detection method using intensity images, the rail detection method provided by the embodiment of the present invention has a smaller rail profile measurement error obtained by fusing images, where the rail profile measurement error is smaller. The maximum measurement error of the maximum value method is reduced to 0.17mm from 0.22mm, the maximum measurement error of the gray scale gravity center method is reduced to 0.14mm from 1.10mm, and the maximum measurement error of the Steger method is reduced to 0.09mm from 0.14 mm.
Compared with the steel rail profile light bar image obtained by the traditional steel rail profile detection method, the fused image obtained by the implementation of the method has no over-exposure phenomenon, the quality of the light bar is obviously improved, the extraction result of the center of the light bar is accurate, the real profile of the steel rail can be better reflected, and the problem of larger error caused by the image over-exposure in the traditional steel rail profile detection method under the complex working condition is better solved.
Based on the same inventive concept, the embodiment of the present invention further provides a rail profile detection system based on polarization splitting, as described in the following embodiments. Because the principle of solving the problems of the embodiment of the system is similar to the method for detecting the steel rail profile based on the polarization beam splitting, the implementation of the embodiment of the system can refer to the implementation of the method, and repeated parts are not repeated.
As shown in fig. 5, a rail profile detection system based on polarization splitting according to an embodiment of the present invention may include: a rail profile detection assembly 5 and an image processing host machine 9. As shown in fig. 4, the rail profile detection assembly 5 may specifically include: the device comprises a line structured light laser 1, a polarization beam splitter 4, a first imaging device (comprising a first imaging lens 3-1 and a first area-array camera 2-1) and a second imaging device (comprising a second imaging lens 3-2 and a second area-array camera 2-2).
Wherein, the line structure light laser 1 emits light knife plane vertically incident to the surface of the steel rail 6; collecting the surface reflected light of the steel rail 6 by a polarization beam splitter prism 4, and decomposing the surface reflected light of the steel rail 6 into P component transmitted light and S component reflected light with mutually vertical vibration planes; p component transmission light is collected through a first imaging lens 3-1, and a P component polarization image is generated according to the P component transmission light collected through the first imaging lens through a first area array camera 2-1 connected with a second imaging lens; s component reflected light is collected through a second imaging lens 3-2, and an S component polarized image is generated according to the S component reflected light collected by the second imaging lens through a second area array camera 2-2 connected with the second imaging lens; the image processing host 9 is in communication with the first area-array camera 2-1 and the second area-array camera 2-2 respectively, determines a total intensity image of the reflected light on the surface of the steel rail according to the P-component polarization image generated by the first imaging device and the S-component polarization image generated by the second imaging device, performs fusion processing on the P-component polarization image, the S-component polarization image and the total intensity image to obtain a fusion image, and extracts the profile data of the steel rail according to the fusion image.
As can be seen from the above, in the rail profile detection system based on polarization splitting provided in the embodiment of the present invention, the polarization splitting prism is used to split the reflected light on the surface of the rail into the P-component transmitted light and the S-component reflected light with the vibration planes perpendicular to each other, the first imaging device generates the P-component polarized image according to the P-component transmitted light, the second imaging device generates the S-component polarized image according to the S-component reflected light, the image processing host determines the total intensity image of the reflected light on the surface of the rail according to the P-component polarized image and the S-component polarized image, and then performs fusion processing on the P-component polarized image, the S-component polarized image and the total intensity image, and the rail profile data is extracted according to the fusion image.
By the rail profile detection system based on the polarization beam splitting, provided by the embodiment of the invention, the P component polarization image, the S component polarization image and the total intensity image of the rail profile light bar are fused, so that a high-quality steel profile light bar image without a local overexposure phenomenon can be obtained, an accurate light bar center is extracted from the high-quality steel profile light bar image, and the rail profile detection accuracy is improved.
In one embodiment, the image processing host 9 may further be configured to: determining the fusion weight of the P component polarization image, the S component polarization image and the total intensity image according to the characteristic information of the light bars in the P component polarization image, the S component polarization image and the total intensity image; and according to the fusion weight of the P component polarization image, the S component polarization image and the total intensity image, carrying out fusion processing on the P component polarization image, the S component polarization image and the total intensity image to obtain a fusion image.
Alternatively, the image processing host computer 9 may determine the fusion weight of the P-component polarization image, the S-component polarization image and the total intensity image by any one of ① determining the fusion weight of the different light bar polarization images according to the width information of the light bars in the P-component polarization image, the S-component polarization image and the total intensity image, and ② determining the fusion weight of the different light bar polarization images according to the intensity information of the light bars in the P-component polarization image, the S-component polarization image and the total intensity image.
Alternatively, the image processing host 9 may perform the fusion process on the P component polarization image, the S component polarization image, and the total intensity image by the above equation (3).
Further, the above-mentioned image processing host 9 may determine the fusion weight of the P component polarization image, the S component polarization image, and the total intensity image by the above-mentioned formula (2).
In one embodiment, the image processing host 9 may further be configured to: extracting the position information of the center of the optical strip in the fusion image; coordinate conversion is carried out on the position information of the center of the optical strip in the fusion image, and the profile data of the half section of one side of the steel rail is obtained; and carrying out profile splicing treatment on the profile data of the half sections on the two sides of the steel rail to obtain the profile data of the full section of the steel rail.
Alternatively, the image processing host 9 may extract the position information of the center of the optical stripe in the fused image by using any one of the following algorithms: the maximum method, the gray scale center of gravity method, and the Steger method.
Based on the same inventive concept, the embodiment of the present invention further provides a rail profile detection apparatus based on polarization splitting, as described in the following embodiments. Because the principle of solving the problems of the embodiment of the device is similar to the method for detecting the steel rail profile based on the polarization beam splitting, the implementation of the embodiment of the device can refer to the implementation of the method, and repeated parts are not repeated.
Fig. 15 is a schematic view of a rail profile detection apparatus based on polarization splitting according to an embodiment of the present invention, and as shown in fig. 15, the apparatus may include: an image acquisition unit 151, a first image processing unit 152, a second image processing unit 153, and a rail profile extraction unit 154.
The image acquiring unit 151 is configured to acquire a P-component polarized image and an S-component polarized image of the reflected light on the surface of the steel rail, where the reflected light on the surface of the steel rail is decomposed into a P-component transmitted light and an S-component reflected light with mutually perpendicular vibration planes through a polarization splitting prism, the P-component polarized image is an image formed by the P-component transmitted light, and the S-component polarized image is an image formed by the S-component reflected light; a first image processing unit 152, configured to determine a total intensity image of the reflected light on the surface of the steel rail according to the P-component polarization image and the S-component polarization image; a second image processing unit 153, configured to perform fusion processing on the P-component polarization image, the S-component polarization image, and the total intensity image to obtain a fusion image; and a rail profile extraction unit 154 for extracting rail profile data from the fused image.
As can be seen from the above, in the polarization-beam-splitting-based steel rail profile detection apparatus provided in the embodiment of the present invention, the image obtaining unit 151 obtains the P-component polarization image and the S-component polarization image of the steel rail profile light bar; determining a total intensity image of the reflected light on the surface of the steel rail according to the P component polarization image and the S component polarization image through the first image processing unit 152; the P-component polarization image, the S-component polarization image, and the total intensity image are fused by the second image processing unit 153, and finally, the rail profile data is extracted from the fused image by the rail profile extraction unit 154.
By the steel rail profile detection device based on the polarization beam splitting, the P component polarization image, the S component polarization image and the total intensity image of the steel rail profile light bar are fused, a high-quality steel rail profile light bar image without a local overexposure phenomenon can be obtained, an accurate light bar center is extracted from the high-quality steel rail profile light bar image, and the accuracy of steel rail profile detection is improved.
In an alternative embodiment, in the rail profile detection apparatus based on polarization splitting provided in the embodiment of the present invention, the second image processing unit 153 may be specifically configured to: determining the fusion weight of the P component polarization image, the S component polarization image and the total intensity image according to the characteristic information of the light bars in the P component polarization image, the S component polarization image and the total intensity image; and according to the fusion weight of the P component polarization image, the S component polarization image and the total intensity image, carrying out fusion processing on the P component polarization image, the S component polarization image and the total intensity image to obtain a fusion image.
Further, the second image processing unit 153 may be further configured to: determining the fusion weight of the polarization images of different light bars according to the width information of the light bars in the P component polarization image, the S component polarization image and the total intensity image; or determining the fusion weight of the polarization images of different light bars according to the intensity information of the light bars in the P component polarization image, the S component polarization image and the total intensity image.
Alternatively, the second image processing unit 153 may be configured to perform fusion processing on the P component polarization image, the S component polarization image, and the total intensity image by the above formula (2).
Further, the second image processing unit 153 may be configured to determine the fusion weight of the P component polarization image, the S component polarization image, and the total intensity image by the above formula (3).
In an alternative embodiment, in the device for detecting a rail profile based on polarization splitting provided in an embodiment of the present invention, the rail profile extracting unit 154 may specifically include: a light bar center extraction module 154-1, configured to extract position information of the light bar center in the fused image; the steel rail half-section profile data determining module 154-2 is used for performing coordinate conversion on position information of the center of the light bar in the fusion image to obtain half-section profile data of one side of the steel rail; and the steel rail full-section profile data determining module 154-3 is used for performing profile splicing treatment on the half-section profile data on the two sides of the steel rail to obtain the full-section profile data of the steel rail.
Alternatively, the light bar center extraction module 154-1 may be configured to extract the position information of the light bar center in the fused image by using any one of the following algorithms: the maximum method, the gray scale center of gravity method, and the Steger method.
Based on the same conception, the embodiment of the invention also provides computer equipment for solving the technical problem that the error of the detection result of the steel rail profile is larger due to the fact that the light bar image acquired by the conventional steel rail profile detection method has a local overexposure phenomenon and the light bar center is difficult to accurately extract.
Based on the same inventive concept, the embodiment of the invention further provides a computer-readable storage medium, which is used for solving the technical problem that the error of the detection result of the steel rail profile is large due to the fact that the light bar image acquired by the conventional steel rail profile detection method has a local overexposure phenomenon and the light bar center is difficult to accurately extract.
In summary, embodiments of the present invention provide a method, a system, an apparatus, a computer device, and a computer-readable storage medium for detecting a rail profile based on polarization beam splitting, based on a polarization beam splitting principle, simultaneously obtain two polarization images of a rail profile with polarization directions perpendicular to each other, obtain a total intensity image from the polarization images, and then fuse the two polarization images and the total intensity image to obtain a fused rail profile light bar image, which can improve the imaging quality of a rail profile light bar, solve the problem of local overexposure of the rail profile light bar image, and improve the rail profile detection accuracy.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above-mentioned embodiments are intended to illustrate the objects, technical solutions and advantages of the present invention in further detail, and it should be understood that the above-mentioned embodiments are only exemplary embodiments of the present invention, and are not intended to limit the scope of the present invention, and any modifications, equivalent substitutions, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.

Claims (12)

1. A method for detecting the profile of a steel rail based on polarization beam splitting is characterized by comprising the following steps:
the method comprises the steps of obtaining a P component polarization image and an S component polarization image of steel rail surface reflected light, wherein the steel rail surface reflected light is decomposed into P component transmission light and S component reflected light which are perpendicular to each other in a vibration plane through a polarization beam splitter prism, the P component polarization image is an image formed by the P component transmission light, and the S component polarization image is an image formed by the S component reflected light;
determining a total intensity image of the reflected light on the surface of the steel rail according to the P component polarization image and the S component polarization image;
performing fusion processing on the P component polarization image, the S component polarization image and the total intensity image to obtain a fusion image;
and extracting the steel rail outline data according to the fused image.
2. The method of claim 1, wherein performing a fusion process on the P-component polarization image, the S-component polarization image, and the total intensity image to obtain a fused image comprises:
determining the fusion weight of the P component polarization image, the S component polarization image and the total intensity image according to the characteristic information of the light bars in the P component polarization image, the S component polarization image and the total intensity image;
and according to the fusion weight of the P component polarization image, the S component polarization image and the total intensity image, carrying out fusion processing on the P component polarization image, the S component polarization image and the total intensity image to obtain a fusion image.
3. The method of claim 2, wherein determining the fusion weight of the P-component polarization image, the S-component polarization image and the total intensity image according to the characteristic information of the light bars in the P-component polarization image, the S-component polarization image and the total intensity image comprises any one of the following:
determining the fusion weight of the polarization images of different light bars according to the width information of the light bars in the P component polarization image, the S component polarization image and the total intensity image;
and determining the fusion weight of the polarization images of different light bars according to the intensity information of the light bars in the P component polarization image, the S component polarization image and the total intensity image.
4. The method of claim 2, wherein the P component polarization image, the S component polarization image, and the total intensity image are subjected to a fusion process by the following formula:
Figure FDA0002401202830000011
wherein, FkjRepresenting the gray value of the jth column and kth row pixels in the fusion image F;
Figure FDA0002401202830000012
representing the gray value of the jth column and kth row of pixels in the ith image to be fused;
Figure FDA0002401202830000013
representing the fusion weight of the jth pixel in the ith image to be fused; n represents the number of images to be fused.
5. The method of claim 3, wherein the fusion weight of the P component polarization image, the S component polarization image and the total intensity image is determined by the following formula:
Figure FDA0002401202830000021
wherein the content of the first and second substances,
Figure FDA0002401202830000022
Figure FDA0002401202830000023
representing the width of a light bar in the jth column in the ith image to be fused; mrRepresenting a reference width of the light bar;
Figure FDA0002401202830000024
representing the difference value between the light strip width of the jth column in the ith image to be fused and the reference width of the light strip; delta0Is shown as
Figure FDA0002401202830000025
A number set when equal to 0 to exclude the denominator being 0; m represents a weight adjustment factor.
6. The method of claim 1, wherein extracting rail profile data from the fused image comprises:
extracting the position information of the center of the optical strip in the fusion image;
coordinate conversion is carried out on the position information of the center of the optical strip in the fusion image, and the profile data of the half section of one side of the steel rail is obtained;
and carrying out profile splicing treatment on the profile data of the half sections on the two sides of the steel rail to obtain the profile data of the full section of the steel rail.
7. The method of claim 6, wherein the position information of the center of the light bar in the fused image is extracted by any one of the following algorithms: the maximum method, the gray scale center of gravity method, and the Steger method.
8. A rail profile detection system based on polarization beam splitting is characterized by comprising: rail profile shape determine module and image processing host computer, rail profile shape determine module includes: the device comprises a line structured light laser, a polarization splitting prism, a first imaging device and a second imaging device;
the line structured light laser is used for emitting a light knife plane to vertically enter the surface of the steel rail;
the polarization beam splitter prism is used for collecting the reflected light on the surface of the steel rail and splitting the reflected light on the surface of the steel rail into P component transmitted light and S component reflected light with mutually vertical vibration planes;
the first imaging device is used for acquiring P component transmission light and generating a P component polarization image;
the second imaging device is used for collecting S-component reflected light and generating an S-component polarized image;
the image processing host is communicated with the first imaging device and the second imaging device respectively, and is used for determining a total intensity image of the reflected light on the surface of the steel rail according to the P component polarization image generated by the first imaging device and the S component polarization image generated by the second imaging device, performing fusion processing on the P component polarization image, the S component polarization image and the total intensity image to obtain a fusion image, and extracting profile data of the steel rail according to the fusion image.
9. The system of claim 8, wherein the first imaging device comprises: a first imaging lens and a first area-array camera; the second imaging device includes: a second imaging lens and a second area-array camera;
the first imaging lens is used for collecting P component transmission light;
the first area-array camera is connected with the first imaging lens and used for generating a P component polarized image according to the P component transmitted light collected by the first imaging lens;
the second imaging lens is used for collecting S component reflected light;
the second area-array camera is connected with the second imaging lens and used for generating an S-component polarized image according to the S-component reflected light collected by the second imaging lens.
10. A rail profile detection device based on polarization beam splitting is characterized by comprising:
the device comprises an image acquisition unit, a polarization splitting prism and a polarization splitting unit, wherein the image acquisition unit is used for acquiring a P component polarization image and an S component polarization image of the reflected light on the surface of the steel rail, the reflected light on the surface of the steel rail is decomposed into a P component transmission light and an S component reflected light which are mutually vertical to a vibration plane through the polarization splitting prism, the P component polarization image is an image formed by the P component transmission light, and the S component polarization image is an image formed by the S component reflected;
the first image processing unit is used for determining a total intensity image of the reflected light on the surface of the steel rail according to the P component polarization image and the S component polarization image;
the second image processing unit is used for carrying out fusion processing on the P component polarization image, the S component polarization image and the total intensity image to obtain a fusion image;
and the steel rail profile extraction unit is used for extracting the steel rail profile data according to the fusion image.
11. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the method for detecting a rail profile based on polarization splitting according to any one of claims 1 to 7 when executing the computer program.
12. A computer-readable storage medium storing a computer program for executing the polarization-beam-splitting-based rail profile detection method according to any one of claims 1 to 7.
CN202010147291.2A 2020-03-05 2020-03-05 Steel rail profile detection method, system and device based on polarization beam splitting Active CN111307065B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010147291.2A CN111307065B (en) 2020-03-05 2020-03-05 Steel rail profile detection method, system and device based on polarization beam splitting

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010147291.2A CN111307065B (en) 2020-03-05 2020-03-05 Steel rail profile detection method, system and device based on polarization beam splitting

Publications (2)

Publication Number Publication Date
CN111307065A true CN111307065A (en) 2020-06-19
CN111307065B CN111307065B (en) 2021-10-26

Family

ID=71149475

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010147291.2A Active CN111307065B (en) 2020-03-05 2020-03-05 Steel rail profile detection method, system and device based on polarization beam splitting

Country Status (1)

Country Link
CN (1) CN111307065B (en)

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103868472A (en) * 2013-12-23 2014-06-18 黑龙江科技大学 Planar structured light three dimension measuring device and method for high-reflectivity part
CN106033599A (en) * 2015-03-20 2016-10-19 南京理工大学 Visible light enhancement method based on polarized imaging
EP3190381A1 (en) * 2014-09-03 2017-07-12 Beijing Jiaotong University System for simultaneously measuring six-degree-of-freedom errors in way that double-frequency lasers are coupled by single optical fiber
CN107633495A (en) * 2017-08-02 2018-01-26 中北大学 A kind of infrared polarization based on complementary relationship and the more embedded fusion methods of algorithm 2D VMD of intensity image
CN108492274A (en) * 2018-04-03 2018-09-04 中国人民解放军国防科技大学 Long-wave infrared polarization feature extraction and fusion image enhancement method
CN109410161A (en) * 2018-10-09 2019-03-01 湖南源信光电科技股份有限公司 A kind of fusion method of the infrared polarization image separated based on YUV and multiple features
CN110388881A (en) * 2019-07-30 2019-10-29 中国铁道科学研究院集团有限公司基础设施检测研究所 Profile of steel rail detection method and device based on polarization imaging
CN111292279A (en) * 2020-01-17 2020-06-16 中国科学院上海技术物理研究所 Polarization image visualization method based on color image fusion

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103868472A (en) * 2013-12-23 2014-06-18 黑龙江科技大学 Planar structured light three dimension measuring device and method for high-reflectivity part
EP3190381A1 (en) * 2014-09-03 2017-07-12 Beijing Jiaotong University System for simultaneously measuring six-degree-of-freedom errors in way that double-frequency lasers are coupled by single optical fiber
CN106033599A (en) * 2015-03-20 2016-10-19 南京理工大学 Visible light enhancement method based on polarized imaging
CN107633495A (en) * 2017-08-02 2018-01-26 中北大学 A kind of infrared polarization based on complementary relationship and the more embedded fusion methods of algorithm 2D VMD of intensity image
CN108492274A (en) * 2018-04-03 2018-09-04 中国人民解放军国防科技大学 Long-wave infrared polarization feature extraction and fusion image enhancement method
CN109410161A (en) * 2018-10-09 2019-03-01 湖南源信光电科技股份有限公司 A kind of fusion method of the infrared polarization image separated based on YUV and multiple features
CN110388881A (en) * 2019-07-30 2019-10-29 中国铁道科学研究院集团有限公司基础设施检测研究所 Profile of steel rail detection method and device based on polarization imaging
CN111292279A (en) * 2020-01-17 2020-06-16 中国科学院上海技术物理研究所 Polarization image visualization method based on color image fusion

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
原惠峰: "不同图像融合技术的融合性能分析比较", 《中国科技论文》 *
王乐: "基于偏振融合的钢轨廓形线结构光成像方法", 《光学学报》 *
王新: "基于偏振成像和图像融合的目标识别技术", 《激光与红外》 *

Also Published As

Publication number Publication date
CN111307065B (en) 2021-10-26

Similar Documents

Publication Publication Date Title
CN108332708B (en) Automatic detection system and detection method for laser level meter
CN106767565A (en) One kind is used for brake block steel back plate plane degree automatic detection device and detection method
CN108801164B (en) Method and system for testing gap value of workpiece based on laser
CN111369484B (en) Rail profile detection method and device
CN104964982A (en) Glass surface authentic and false defect identification method and system based on OCT complex signal
NL2016637B1 (en) Rail profile monitoring, e.g. geometry of the frogs.
CN107561091A (en) A kind of detecting system and detection method of oblique fire formula solid wood board face crack
CN106996744A (en) A kind of real-time displacement device for accurately measuring and measuring method based on bar code ruler
CN106767545A (en) A kind of high accuracy high-space resolution angel measuring instrument and angle measurement method
CN104554344A (en) Tread defect information detection system and method
CN105160754A (en) Coin surface quality detection apparatus and coin surface quality detection method based on height measurement
CN105066903A (en) Laser three-dimensional measurement system and measurement method thereof
KR20160102244A (en) Non-imaging coherent line scanner systems and methods for optical inspection
CN108981618A (en) A kind of tunnel just branch flatness analysis method and system
CN100339680C (en) T-shape guide linear laser detecting instrument
Bursanescu et al. Automated pavement distress data collection and analysis: a 3-D approach
KR102513815B1 (en) Method and device for detecting abnormalities in train tracks based on automation
US20030210406A1 (en) Sensor alignment method for 3D measurment systems
CN111369533B (en) Rail profile detection method and device based on polarization image fusion
CN111307065B (en) Steel rail profile detection method, system and device based on polarization beam splitting
CN108225187B (en) Wavefront sensing-based aspheric lens error detection method
CN111336939B (en) Machine vision-based online detection method and device for gap between explosion-proof joint surfaces
CN103884294A (en) Wide view field type infrared light three-dimensional morphology measurement method and device thereof
KR101865434B1 (en) Method and evaluation device for determining the position of a structure located in an object to be examined by means of x-ray computer tomography
US10091493B2 (en) Device and method for scanning object outline image

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant