CN114549389A - Wheel set tread detection method, device, system, terminal and storage medium - Google Patents
Wheel set tread detection method, device, system, terminal and storage medium Download PDFInfo
- Publication number
- CN114549389A CN114549389A CN202011292110.1A CN202011292110A CN114549389A CN 114549389 A CN114549389 A CN 114549389A CN 202011292110 A CN202011292110 A CN 202011292110A CN 114549389 A CN114549389 A CN 114549389A
- Authority
- CN
- China
- Prior art keywords
- point
- point cloud
- wheel set
- tread
- train wheel
- 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.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 54
- 238000003860 storage Methods 0.000 title claims abstract description 15
- 230000002159 abnormal effect Effects 0.000 claims abstract description 19
- 238000000034 method Methods 0.000 claims description 65
- 238000012545 processing Methods 0.000 claims description 31
- 238000004590 computer program Methods 0.000 claims description 20
- 239000011159 matrix material Substances 0.000 claims description 19
- 230000006870 function Effects 0.000 claims description 16
- 238000001914 filtration Methods 0.000 claims description 15
- 238000013519 translation Methods 0.000 claims description 10
- 238000013507 mapping Methods 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 7
- 230000001131 transforming effect Effects 0.000 claims description 4
- 238000005516 engineering process Methods 0.000 abstract description 4
- 238000010586 diagram Methods 0.000 description 15
- 238000007689 inspection Methods 0.000 description 8
- 230000008569 process Effects 0.000 description 8
- 238000012986 modification Methods 0.000 description 5
- 230000004048 modification Effects 0.000 description 5
- 238000005488 sandblasting Methods 0.000 description 5
- 230000008901 benefit Effects 0.000 description 3
- 208000037656 Respiratory Sounds Diseases 0.000 description 2
- 206010039203 Road traffic accident Diseases 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000014509 gene expression Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000011179 visual inspection Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/20—Image enhancement or restoration using local operators
- G06T5/30—Erosion or dilatation, e.g. thinning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- 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
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
- G06T2207/10012—Stereo images
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Quality & Reliability (AREA)
- Image Processing (AREA)
Abstract
The embodiment of the application provides a wheel set tread detection method, device, system, terminal and storage medium, and relates to a quality detection technology. The wheel set tread detection method comprises the following steps: denoising the acquired point cloud of the train wheel set to obtain a tread point cloud of the train wheel set; registering the obtained tread point cloud with the tread point cloud in the standard map; and when the distance between the point in the registered point cloud and the nearest point in the standard map is larger than a first threshold value, determining the point as an abnormal point.
Description
Technical Field
The present disclosure relates to quality detection technologies, and in particular, to a method, an apparatus, a system, a terminal, and a storage medium for detecting wheel set tread.
Background
In the related technology, the detection scheme of the high-speed rail wheel on the tread crack adopts manual detection, an inspection worker needs to go into an inspection channel below the high-speed rail to be inspected, irradiates an area to be inspected with a flashlight and records the area on a paper document to form an inspection record and a report document.
However, as the number of high-speed rail vehicles is continuously increased, the labor cost of the manual detection scheme is increased linearly, and the inspection pressure of workers is greatly increased. The detection precision of workers can be obviously reduced under the condition of overlong working length, and once the cracks on the wheel set tread are not detected, serious traffic accidents can occur, and driving safety is not facilitated. Furthermore, the scheme of manual detection cannot guarantee that a worker can detect each point position, hidden dangers are buried for the quality of high-speed rails, and driving safety is not facilitated.
Disclosure of Invention
In order to solve one of the technical defects, embodiments of the present application provide a wheel set tread detection method, device, system, terminal, and storage medium.
An embodiment of a first aspect of the present application provides a wheel set tread detection method, including:
denoising the acquired point cloud of the train wheel set to obtain a tread point cloud of the train wheel set;
registering the obtained tread point cloud with the tread point cloud in the standard map;
and when the distance between the point in the registered point cloud and the nearest point in the standard map is larger than a first threshold value, determining the point as an abnormal point.
An embodiment of a second aspect of the present application provides a wheel set tread surface detection device, including:
the acquisition module is used for carrying out denoising treatment on the acquired point cloud of the train wheel set to obtain a tread point cloud of the train wheel set;
the first processing module is used for registering the obtained tread point cloud with the tread point cloud in the standard map;
and the second processing module is used for determining the point as an abnormal point when the distance between the point in the registered point cloud and the nearest point in the standard map is greater than a first threshold value.
An embodiment of a third aspect of the present application provides a wheel set tread surface detection system, including:
wheel-to-tread detection apparatus as claimed in any one of the preceding claims;
the robot is provided with a camera device for acquiring image data of the train wheel set tread and is used for sending the image data to the wheel set tread detection device so that the wheel set tread detection device can acquire point cloud of the train wheel set according to the image data.
An embodiment of a fourth aspect of the present application provides a terminal, including:
a memory; the device can support a processor to read original point clouds, and simultaneously support the processor to store data processed by any one of the methods.
A processor; the method can read in original point cloud from a memory, and process the point cloud according to any one of the methods 1-8 to obtain processed point cloud data and image data.
A computer program; the method of any one of the preceding 1-8 can be implemented in a computer language, compiled, and run quickly on a processor, stored in the memory.
An embodiment of a fifth aspect of the present application provides a computer-readable storage medium having a computer program stored thereon; the computer program is executed by a processor to implement a method as claimed in any preceding claim.
The embodiment of the application provides a wheel set tread detection method, a device, a system, a terminal and a storage medium, wherein a point cloud of a current wheel set is obtained through acquisition, other devices except for the wheel set in the point cloud are filtered out, only the point cloud of the wheel set is reserved, the point cloud of the current wheel set is aligned with the point cloud of a wheel set in a standard map, and then the point cloud of the current wheel set is compared with the point cloud of the wheel set in the standard map, so that abnormal points are screened out. Therefore, the detection of the wheel pair tread can be automatically, quickly and accurately realized, the labor is saved, and the detection precision and reliability are improved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:
FIG. 1 is a schematic flow chart of a method provided in an exemplary embodiment;
FIG. 2 is a schematic flow diagram of a method provided by another exemplary embodiment;
FIG. 3 is a block diagram of an apparatus provided in an exemplary embodiment;
fig. 4 is a block diagram of a system provided in an exemplary embodiment.
Detailed Description
In order to make the technical solutions and advantages of the embodiments of the present application more apparent, the following further detailed description of the exemplary embodiments of the present application with reference to the accompanying drawings makes it clear that the described embodiments are only a part of the embodiments of the present application, and are not exhaustive of all embodiments. It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict.
In the related technology, the detection scheme of the high-speed rail wheel on the tread crack adopts manual detection, an inspection worker needs to go into an inspection channel below the high-speed rail to be inspected, irradiates an area to be inspected with a flashlight and records the area on a paper document to form an inspection record and a report document.
However, as the number of high-speed rail vehicles is continuously increased, the labor cost of the manual detection scheme is increased linearly, and the inspection pressure of workers is increased greatly. The detection precision of workers can be obviously reduced under the condition of overlong working length, and once the cracks on the wheel set tread are not detected, serious traffic accidents can occur, and driving safety is not facilitated. Furthermore, the scheme of manual detection cannot guarantee that a worker can detect each point position, hidden dangers are buried for the quality of high-speed rails, and driving safety is not facilitated.
In order to overcome the problems, the embodiment of the application provides a wheel set tread detection method, a device, a system, a terminal and a storage medium, wherein point cloud of a current wheel set is obtained through acquisition, other devices except the wheel set in the point cloud are filtered, and only the point cloud of the wheel set is reserved; meanwhile, aligning the current wheel set point cloud with the wheel set point cloud of the standard map, comparing the two point clouds, and screening abnormal points; further, the problem of false detection points caused by the edge of the wheel set due to different widths of the wheel set needs to be solved, point clouds of wheel set tread cracks are reserved, and the point clouds are mapped back to a 2D image. Therefore, the crack of the wheel set tread can be automatically, quickly and accurately identified, labor is saved, and detection precision and reliability are improved.
The functions and implementation of the method provided by the present embodiment are described below with reference to the accompanying drawings.
As shown in fig. 1, the wheel set tread detecting method provided by this embodiment includes:
s101, denoising the acquired point cloud of the train wheel set to obtain a tread point cloud of the train wheel set;
s102, registering the obtained tread point cloud and the tread point cloud in the standard graph;
s103, when the distance between the point in the registered point cloud and the nearest point in the standard map is larger than a first threshold value, determining that the point is an abnormal point.
Before step S101, the tread of the train wheel set needs to be collected to obtain a depth map of the wheel set. Specifically, wheel set tread data acquired by a camera device of the robot is acquired; and acquiring a depth map of the item points to be detected according to wheel set tread data, and obtaining corresponding point clouds through camera internal parameter calculation. The robot sends stepping data acquired by the camera to terminals such as a server for analysis, and the whole process does not exceed 1 s; and a plurality of wheel sets of the train can be acquired and analyzed simultaneously, so that the detection time is further shortened.
In step S101, due to the shooting angle, the current frame point cloud generally includes devices near the wheel set, such as parts of rails and the ground, a sand blasting gun, a fixing bolt, and the like, in addition to the point cloud of the wheel set to be detected. There are standard drawings without these parts, and the extra parts at different wheel positions are different, so the extra parts need to be filtered out, only the tread part is reserved, and the subsequent matching is convenient.
Specifically, carry out the processing of removing noise to the point cloud of the train wheel pair who obtains, include:
carrying out deep low-pass filtering processing on the acquired point cloud of the train wheel set;
and clustering the filtered point cloud.
In the depth low-pass filtering stage, the depth of a corresponding point can be obtained according to a depth map of a wheel pair obtained in advance; through a depth low-pass filtering method, namely, the depth of the acquired corresponding point passes through a low-pass filter with a fixed threshold value, other parts except the sand blasting gun can be removed, and the method comprises the following specific steps:
where d (i) is the depth of the ith point, and c is the distance threshold.
After the depth low-pass filtering method is adopted, only wheel set tread point cloud and sand blasting gun point cloud are left in the point cloud, and the point cloud is close in depth and cannot be removed through depth low-pass filtering. Therefore, two point clouds need to be segmented by a clustering method. Since the point clouds of the two devices are continuous in shape, the Clustering method adopts a DBSCAN (English-Based Spatial Clustering of Applications with Noise) Clustering method.
Specifically, DBSCAN requires two parameters: the scan radius (eps) and the minimum contained number of points (minPts). Optionally, starting with a point that is not visited (unvisited), find all nearby points that are within eps distance (including eps) from it. If the number of nearby points is ≧ minPts, the current point forms a cluster with its nearby points, and the departure point is marked as visited (visited). Then recursively, all points in the cluster that are not marked as accessed (visited) are processed in the same way, thereby expanding the cluster. If the number of nearby points < minPts, the point is temporarily marked as a noise point. If the cluster is sufficiently expanded, i.e., all points within the cluster are marked as visited, then the same algorithm is used to process the points that are not visited.
In step S102, 3D point cloud registration is performed on the denoised point cloud: and registering the wheel set tread Point cloud obtained in the step with the wheel set tread Point cloud of the standard graph, wherein the Point-to-Point Closest Point matching (ICP) is adopted in the registering method.
Specifically, a second point set corresponding to the first point set is determined from the target point cloud, wherein points in the second point set are the closest points in the first point set, and the first point set is obtained from the source point cloud;
acquiring a rotation matrix and a translation matrix which can minimize an error function;
transforming the points in the first point set according to the rotation matrix and the translation matrix to obtain a third point set;
and determining the average distance between the third point set and the fourth point set, and determining that the registration is finished when the average distance is smaller than a second threshold or the iteration number reaches a third threshold.
Wherein the target point cloud is the obtained tread point cloud; the source point cloud is the point cloud before denoising.
For example:
step 1: taking a point set pi from a source point cloud P, wherein the pi belongs to P;
step 2: finding out a corresponding point set qi in the target point cloud Q, wherein qi belongs to Q, and min is obtained from qi to pi;
and step 3: calculating a rotation matrix R and a translation matrix t to minimize an error function;
and 4, step 4: carrying out rotation and translation transformation on pi by using R and t obtained in the previous step to obtain a new corresponding point set
p′i＝{p′i＝Rpi+t,pi∈P}
And 5: calculating the average distance d between p 'i and the corresponding point set qi as 1n sigma ni as 1/p' i-qi/2;
step 6: if d is smaller than a given threshold or larger than a preset maximum iteration number, stopping iterative computation, otherwise, returning to the step 2.
After the iterative calculations are complete, then the anomaly points for the tread surface can be determined. After the 3D point cloud registration is completed, the point cloud of the current wheel set tread and the point cloud of the standard map tread are basically overlapped in space, and the distance from each point in the current point cloud to the nearest point on the standard map can be obtained. If the closest distance exceeds a certain threshold, it is considered as an outlier.
Because the abnormal point includes the wheel pair edge because the false positive point that the wheel pair width problem caused to and the point of true crackle, in order to further improve the accuracy nature that the crackle detected, still need further filtering the abnormal point, with filtering the false positive point in the abnormal point. When the false alarm points are filtered, the false alarm points at the edge are filtered through a mask of the wheel set outline acquired in advance, the point cloud of the real crack is obtained, and then the internal parameters of the camera are mapped back to the 2D plane, so that the accurate positioning of the tread crack is realized.
Specifically, after obtaining the tread point cloud of the train wheel set, the method further comprises the following steps:
mapping the tread point cloud of the train wheel set to a depth map;
detecting an outer frame of the train wheel set on the depth map;
and carrying out expansion treatment on the outer frame of the train wheel set to obtain the contour mask of the train wheel set.
Correspondingly, when the distance between the point in the registered point cloud and the nearest point in the standard map is greater than a first threshold, after determining that the point is an abnormal point, the method further comprises:
filtering false alarm points in the abnormal points through a contour mask of the train wheel set to obtain points at the crack;
the point at the crack is mapped to the 2D plane using camera parameters.
After mapping the point at the crack to the 2D plane, displaying may be performed at a corresponding terminal or display device; in the display interface, the color or line type and the like at the crack can be different from other areas so as to facilitate the visual inspection of a worker.
In each of the above examples, the specific value of each threshold may be set according to actual needs, and this embodiment is not particularly limited.
According to the wheel set tread detection method provided by the embodiment, the point cloud of the current wheel set is obtained, other devices except for the wheel set in the point cloud are filtered, and only the point cloud of the wheel set is reserved; meanwhile, aligning the current wheel set point cloud with the wheel set point cloud of the standard map, comparing the two point clouds, and screening abnormal points; further, the problem of false detection points caused by the edge of the wheel set due to different widths of the wheel set needs to be solved, point clouds of wheel set tread cracks are reserved, and the point clouds are mapped back to a 2D image. Therefore, the crack of the wheel set tread can be automatically, quickly and accurately identified, labor is saved, and detection precision and reliability are improved.
In a more preferred example, as shown in fig. 2, the method of this embodiment may include the following steps:
obtaining a current wheel pair depth map: reading a depth map of the item points to be detected, and obtaining corresponding point clouds for subsequent processing and calculation through camera internal parameter calculation;
depth low-pass filtering: the step is used for removing devices at the background of the point cloud of the wheel set, such as parts of rails, the ground, fixing bolts and the like, and only the tread and the sand blasting gun are reserved, so that subsequent matching is facilitated;
wheel set point cloud clustering: the step is used for removing the point cloud of the sand blasting gun which is very close to the wheel set in the point cloud, and only the point cloud of the tread is reserved, so that the subsequent matching is facilitated;
wheel set outline extraction: the step is used for extracting the edge profile of the wheel set, generating a mask for subsequently removing false detection caused by the width problem of the wheel set;
3D point cloud registration: the step is used for calculating a point cloud transformation matrix from the current wheel set to the standard map wheel set to realize the point-to-point alignment of two point clouds;
positioning tread cracks: the method comprises the steps of calculating abnormal points in current tread surface point clouds, removing false detection points on the edge of a tread surface by using a mask, retaining the tread surface crack point clouds, mapping the tread surface crack point clouds to a 2D picture, and realizing positioning of the tread surface cracks.
The method provided by the embodiment can be particularly used for detecting the tread of the wheel set of the rail vehicle such as a high-speed rail.
Compared with a manual detection scheme in the related art, the method provided by the embodiment has the advantage that the detection speed is obviously improved. The wheel set tread is collected through the method that the robot is additionally provided with the camera to collect to data transmission that will gather carries out the analysis to terminals such as server, and whole process is no longer than 1s, and can go on simultaneously to a plurality of wheel sets of train, further reduces check-out time, improves detection efficiency.
Compared with the manual detection scheme in the related art, the method provided by the embodiment can avoid errors in manual detection, and the detection precision is obviously improved. The embodiment can position the position of the tread by utilizing the depth map information, and then accurately detect the position of a crack on the tread of the wheel set.
Compared with the manual detection scheme in the related art, the method provided by the embodiment has the advantage that the detection reliability is remarkably improved. According to the embodiment, the robot is used for collecting the pictures, the server and other terminals are used for analyzing and processing, the whole process can be fully automated, the operation is carried out 24 hours all day, the reliability of the device is higher compared with that of manual long-time detection, and the detection efficiency can be further improved.
The embodiment also provides a wheel set tread detection device. The wheel set tread detecting device provided by this embodiment is a product embodiment corresponding to the method embodiment, and the function and implementation process thereof are the same as or similar to those of the previous embodiment, and details are not repeated here.
As shown in fig. 3, the wheel-set tread detecting device provided by this embodiment includes:
the acquisition module 11 is configured to perform denoising processing on the acquired point cloud of the train wheel set to obtain a tread point cloud of the train wheel set;
the first processing module 12 is used for registering the obtained tread point cloud with the tread point cloud in the standard map;
and a second processing module 13, configured to determine that a point is an outlier when a distance between a point in the registered point cloud and a closest point in the standard map is greater than a first threshold.
In one possible implementation manner, the obtaining unit 11 is specifically configured to:
carrying out deep low-pass filtering processing on the acquired point cloud of the train wheel set;
and clustering the filtered point cloud.
In one possible implementation manner, the obtaining unit 11 is specifically configured to:
and processing the filtered point cloud by adopting a DBSCAN clustering method.
In one possible implementation manner, the obtaining unit 11 is further configured to:
acquiring wheel set tread data acquired by a camera device of a robot;
and acquiring a depth map of the item points to be detected according to wheel set tread data, and obtaining corresponding point clouds through camera internal parameter calculation.
In one possible implementation manner, the first processing module 12 is specifically configured to:
and registering the obtained tread point cloud and the tread point cloud in the standard graph by adopting a point-to-point closest point matching method.
In one possible implementation manner, the first processing module 12 is specifically configured to:
determining a second point set corresponding to the first point set from the target point cloud, wherein points in the second point set are the closest points in the first point set, and the first point set is obtained from the source point cloud;
acquiring a rotation matrix and a translation matrix which can minimize an error function;
transforming the points in the first point set according to the rotation matrix and the translation matrix to obtain a third point set;
and determining the average distance between the third point set and the fourth point set, and determining that the registration is finished when the average distance is smaller than a second threshold or the iteration number reaches a third threshold.
In one possible implementation manner, the second processing unit 13 is further configured to:
mapping the tread point cloud of the train wheel set to a depth map;
detecting an outer frame of the train wheel set on the depth map;
and carrying out expansion treatment on the outer frame of the train wheel set to obtain the contour mask of the train wheel set.
In one possible implementation manner, the second processing unit 13 is further configured to:
filtering false alarm points in the abnormal points through a contour mask of the train wheel set to obtain points at the crack;
the point at the crack is mapped to the 2D plane using camera parameters.
As shown in fig. 4, this embodiment further provides a wheel set tread surface detection system, including:
the wheel-set tread detecting device 1 in any one of the previous embodiments;
and the robot 2 is provided with a camera device for acquiring image data of the train wheel set tread and used for sending the image data to the wheel set tread detection device 1 so as to enable the wheel set tread detection device 1 to acquire point cloud of the train wheel set according to the image data.
In a specific implementation, the number of robots in the system may be set according to actual needs, and the embodiment is not specifically limited herein. In addition, the structure of the robot in this example is not particularly limited in this embodiment, as long as the corresponding functions can be realized.
The present embodiment provides a terminal, including:
a memory; the device can support the processor to read the original point cloud, and simultaneously, the processor stores the data processed by the steps.
A processor; the original point cloud can be read in from the memory, and the point cloud is processed according to the steps to obtain processed point cloud data and image data.
A computer program; the complete algorithm function of the steps can be realized through a computer language, the compiling is completed, and the method can be quickly operated in a processor.
Wherein the computer program is stored in a memory and configured to be executed by a processor to implement the respective method.
The memory is used for storing a computer program, and the processor executes the computer program after receiving the execution instruction, and the method executed by the apparatus defined by the flow process disclosed in the foregoing corresponding embodiments can be applied to or implemented by the processor.
The Memory may comprise a Random Access Memory (RAM) and may also include a non-volatile Memory, such as at least one disk Memory. The memory can implement communication connection between the system network element and at least one other network element through at least one communication interface (which may be wired or wireless), and the internet, a wide area network, a local network, a metropolitan area network, and the like can be used.
The processor may be an integrated circuit chip having signal processing capabilities. In implementation, the steps of the method disclosed in the first embodiment may be implemented by hardware integrated logic circuits in a processor or instructions in the form of software. The Processor may be a general-purpose Processor, and includes a Central Processing Unit (CPU), a Network Processor (NP), and the like; but may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components. The corresponding methods, steps, and logic blocks disclosed in the embodiments of the present invention may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The steps of the method disclosed in connection with the embodiments of the present invention may be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software elements in the decoding processor. The software elements may be located in ram, flash, rom, prom, or eprom, registers, among other storage media that are well known in the art. The storage medium is located in a memory, and a processor reads information in the memory and completes the steps of the method in combination with hardware of the processor.
The present embodiment provides a computer-readable storage medium having stored thereon a computer program; the computer program is executed by a processor in a corresponding method. For specific implementation, reference may be made to the method embodiments, which are not described herein again.
It should be noted that: unless specifically stated otherwise, the relative steps, numerical expressions, and values of the components and steps set forth in these embodiments do not limit the scope of the present invention. In all examples shown and described herein, unless otherwise specified, any particular value should be construed as merely illustrative, and not restrictive, and thus other examples of example embodiments may have different values.
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a unit, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application 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 application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. 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.
While the preferred embodiments of the present application have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all alterations and modifications as fall within the scope of the application.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present application without departing from the spirit and scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is intended to include such modifications and variations as well.
Claims (16)
1. A wheel set tread detection method is characterized by comprising the following steps:
denoising the acquired point cloud of the train wheel set to obtain a tread point cloud of the train wheel set;
registering the obtained tread point cloud with the tread point cloud in the standard map;
and when the distance between the point in the registered point cloud and the nearest point in the standard map is larger than a first threshold value, determining the point as an abnormal point.
2. The method of claim 1, wherein the denoising the acquired point cloud of the train wheel set comprises:
carrying out deep low-pass filtering processing on the acquired point cloud of the train wheel set;
and clustering the filtered point cloud.
3. The method of claim 2, wherein clustering the filtered point cloud comprises:
and processing the filtered point cloud by adopting a DBSCAN clustering method.
4. The method of claim 1, wherein before denoising the acquired point cloud of the train wheel set, the method further comprises:
acquiring wheel set tread data acquired by a camera device of a robot;
and acquiring a depth map of the item points to be detected according to the wheel set tread data, and obtaining corresponding point clouds through camera internal parameter calculation.
5. The method of claim 1, wherein registering the obtained tread surface point cloud with the tread surface point cloud in the standard map comprises:
and registering the obtained tread point cloud and the tread point cloud in the standard graph by adopting a point-to-point closest point matching method.
6. The method of claim 5, wherein registering the obtained tread surface point cloud with the tread surface point cloud in the standard map using a point-to-point nearest point matching method comprises:
determining a second point set corresponding to the first point set from the target point cloud, wherein points in the second point set are the closest points in the first point set, and the first point set is obtained from the source point cloud;
acquiring a rotation matrix and a translation matrix which can minimize an error function;
transforming the points in the first point set according to the rotation matrix and the translation matrix to obtain a third point set;
and determining the average distance between the third point set and the fourth point set, and determining that the registration is finished when the average distance is smaller than a second threshold or the iteration number reaches a third threshold.
7. The method of claim 1, further comprising, after obtaining the tread point cloud for the train wheel set:
mapping the tread point cloud of the train wheel set to a depth map;
detecting an outer frame of the train wheel set on the depth map;
and carrying out expansion treatment on the outer frame of the train wheel set to obtain the contour mask of the train wheel set.
8. The method of claim 7, wherein after determining that the point in the registered point cloud is an abnormal point when the distance between the point and the closest point in the standard map is greater than a first threshold, the method further comprises:
filtering false alarm points in the abnormal points through a contour mask of the train wheel set to obtain points at crack positions;
mapping the point at the crack to a 2D plane using camera parameters.
9. A wheel set tread detection device, comprising:
the acquisition module is used for carrying out denoising treatment on the acquired point cloud of the train wheel set to obtain a tread point cloud of the train wheel set;
the first processing module is used for registering the obtained tread point cloud with the tread point cloud in the standard map;
and the second processing module is used for determining the point as an abnormal point when the distance between the point in the registered point cloud and the nearest point in the standard map is greater than a first threshold value.
10. The apparatus according to claim 9, wherein the obtaining unit is specifically configured to:
carrying out deep low-pass filtering processing on the acquired point cloud of the train wheel set;
and clustering the filtered point cloud.
11. The apparatus of claim 9, wherein the obtaining unit is further configured to:
before the point cloud of train wheel pair that obtains carries out the processing of denoising, still include:
acquiring wheel set tread data acquired by a camera device of the robot;
and acquiring a depth map of the item points to be detected according to the wheel set tread data, and obtaining corresponding point clouds through camera internal parameter calculation.
12. The apparatus of claim 9, wherein the first processing module is specifically configured to:
determining a second point set corresponding to the first point set from the target point cloud, wherein points in the second point set are the closest points in the first point set, and the first point set is obtained from the source point cloud;
acquiring a rotation matrix and a translation matrix which can minimize an error function;
transforming the points in the first point set according to the rotation matrix and the translation matrix to obtain a third point set;
and determining the average distance between the third point set and the fourth point set, and determining that the registration is finished when the average distance is smaller than a second threshold or the iteration number reaches a third threshold.
13. The apparatus of claim 12, wherein the second processing unit is further configured to:
mapping the tread point cloud of the train wheel set to a depth map;
detecting an outer frame of the train wheel set on the depth map;
performing expansion treatment on the outer frame of the train wheel set to obtain a contour mask of the train wheel set;
the second processing unit is further configured to:
filtering false alarm points in the abnormal points through a contour mask of the train wheel set to obtain points at crack positions;
mapping the point at the crack to a 2D plane using camera parameters.
14. A wheel set tread detection system, comprising:
wheel-to-tread detecting apparatus as claimed in any one of claims 9 to 13;
the robot is provided with a camera device for acquiring image data of the train wheel set tread and is used for sending the image data to the wheel set tread detection device so that the wheel set tread detection device can acquire point cloud of the train wheel set according to the image data.
15. A terminal, comprising:
a memory; a device capable of supporting a processor to read an original point cloud while storing data processed by the method of any one of claims 1-8;
a processor; reading in original point clouds from a memory and processing the point clouds according to the method of any one of claims 1-8 to obtain processed point cloud data and image data;
a computer program; stored in said memory, capable of implementing the method of any one of claims 1 to 8 in a computer language, performing the compilation and being fast running in the processor.
16. A computer-readable storage medium, having stored thereon a computer program; the computer program is executed by a processor to implement the method of any one of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011292110.1A CN114549389A (en) | 2020-11-18 | 2020-11-18 | Wheel set tread detection method, device, system, terminal and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011292110.1A CN114549389A (en) | 2020-11-18 | 2020-11-18 | Wheel set tread detection method, device, system, terminal and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114549389A true CN114549389A (en) | 2022-05-27 |
Family
ID=81659172
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011292110.1A Pending CN114549389A (en) | 2020-11-18 | 2020-11-18 | Wheel set tread detection method, device, system, terminal and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114549389A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115439480A (en) * | 2022-11-09 | 2022-12-06 | 成都运达科技股份有限公司 | Bolt abnormity detection method and system based on 3D depth image template matching |
| CN117894015A (en) * | 2024-03-15 | 2024-04-16 | 浙江华是科技股份有限公司 | Point cloud annotation data optimization method and system |
-
2020
- 2020-11-18 CN CN202011292110.1A patent/CN114549389A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115439480A (en) * | 2022-11-09 | 2022-12-06 | 成都运达科技股份有限公司 | Bolt abnormity detection method and system based on 3D depth image template matching |
| CN115439480B (en) * | 2022-11-09 | 2023-02-28 | 成都运达科技股份有限公司 | Bolt abnormity detection method and system based on 3D depth image template matching |
| CN117894015A (en) * | 2024-03-15 | 2024-04-16 | 浙江华是科技股份有限公司 | Point cloud annotation data optimization method and system |
| CN117894015B (en) * | 2024-03-15 | 2024-05-24 | 浙江华是科技股份有限公司 | Point cloud annotation data optimization method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107341453B (en) | Lane line extraction method and device | |
| US9365217B2 (en) | Mobile pothole detection system and method | |
| CN107239742B (en) | Method for calculating scale value of instrument pointer | |
| CN109373921B (en) | Tunnel monitoring method and device | |
| CN114549389A (en) | Wheel set tread detection method, device, system, terminal and storage medium | |
| CN110986820B (en) | Track intrusion detection method and device and electronic equipment | |
| KR20020077420A (en) | Method for automatically detecting casting defects in a test piece | |
| US8121415B2 (en) | Combining feature boundaries | |
| CN112365461A (en) | Fastener loosening identification method, system, terminal and storage medium | |
| CN109916488B (en) | Dynamic vehicle weighing method and device | |
| CN112598922B (en) | Parking space detection method, device, equipment and storage medium | |
| US20200279395A1 (en) | Method and system for enhanced sensing capabilities for vehicles | |
| CN112488995B (en) | Intelligent damage judging method and system for automatic maintenance of train | |
| CN117079219A (en) | Vehicle running condition monitoring method and device applied to trailer service | |
| Dhiman et al. | A multi-frame stereo vision-based road profiling technique for distress analysis | |
| Ryu et al. | Feature-based pothole detection in two-dimensional images | |
| CN114332349A (en) | Binocular structured light edge reconstruction method and system and storage medium | |
| CN111488762A (en) | Lane-level positioning method and device and positioning equipment | |
| CN111460198B (en) | Picture timestamp auditing method and device | |
| CN117058222A (en) | TBM rock slag form identification method and system | |
| CN116681912A (en) | Rail gauge detection method and device for railway turnout | |
| CN115761613B (en) | Automatic tunnel crack detection method based on convolutional network | |
| Danilescu et al. | Road anomalies detection using basic morphological algorithms | |
| CN112598314B (en) | Method, device, equipment and medium for determining perception confidence of intelligent driving automobile | |
| CN115331193A (en) | Parking space identification method, parking space identification system, electronic equipment and storage medium |
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 |