CN107330931B - A kind of longitudinal displacement of steel rail detection method and system based on image sequence - Google Patents
A kind of longitudinal displacement of steel rail detection method and system based on image sequence Download PDFInfo
- Publication number
- CN107330931B CN107330931B CN201710388790.9A CN201710388790A CN107330931B CN 107330931 B CN107330931 B CN 107330931B CN 201710388790 A CN201710388790 A CN 201710388790A CN 107330931 B CN107330931 B CN 107330931B
- Authority
- CN
- China
- Prior art keywords
- steel rail
- longitudinal displacement
- image
- detection
- coding maker
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/50—Depth or shape recovery
- G06T7/55—Depth or shape recovery from multiple images
-
- 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
-
- 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
- G06T2207/10016—Video; Image sequence
-
- 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
- G06T2207/20081—Training; Learning
-
- 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
- G06T2207/20084—Artificial neural networks [ANN]
-
- 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
- G06T2207/30204—Marker
-
- 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
- G06T2207/30236—Traffic on road, railway or crossing
Abstract
The present invention discloses a kind of longitudinal displacement of steel rail detection method based on image sequence, which comprises S1: coding maker is arranged on the fixed reference benchmark architecture and steel rail web beside rail;S2: different time, different angle, the longitudinal displacement of steel rail route live image comprising all coding makers are acquired by image capture device and form image sequence;S3: the marker detection method based on convolutional neural networks, the detection of building coding maker and location model are detected and are positioned to the coding maker in image sequence;S4: the characteristic point of each coding maker and the sub-pix image coordinate of characteristic point are decoded;S5: the Three-dimension Reconstruction Model based on sub-pix image coordinate building longitudinal displacement of steel rail route scene;S6: calculating longitudinal displacement of steel rail based on the Three-dimension Reconstruction Model, and the invention also discloses the system using this method, the present invention improves the detection efficiency and accuracy of longitudinal displacement of steel rail detection, and easy to operate, easy to implement.
Description
Technical field
The present invention relates to railway construction technical fields.More particularly, to a kind of rail longitudinal direction position based on image sequence
Move detection method and system.
Background technique
The high ride of track is to guarantee the most important condition of bullet train operation security and comfort, and the height of track is smooth
Property depend on rail geometry state, it is therefore, extremely important to the detection of rail geometry state.Due to expanding with heat and contract with cold and train row
Longitudinal resistance in, rail will appear length travel to a certain extent, are easy to cause expansion rail track or brittle fractures of rail, exist
Huge security risk.
As train running speed is constantly accelerated, train quantity is continuously increased, and the detection work of rail geometry state is increasingly
It is heavy.The method of artificial detection longitudinal displacement of steel rail amount at this stage is complicated for operation, time-consuming, precision is low, and needs early period
Fixed device is set on the line, buries detection device in trackside, occupies a large amount of manpower and material resources, influence route to a certain extent
Operation, is unable to satisfy the requirement of modern railways, the automatic detection of fast accurate has become inevitable development trend.
Accordingly, it is desirable to provide a kind of simple, quick, safety and the method for accurately carrying out longitudinal displacement of steel rail detection, with
Solve the problems, such as that artificial detection longitudinal displacement of steel rail method is complicated for operation, time-consuming at this stage, precision is low.
Summary of the invention
The longitudinal displacement of steel rail detection method based on image sequence that it is an object of the present invention to provide a kind of.The present invention
Another be designed to provide a kind of longitudinal displacement of steel rail detection system based on image sequence, to improve longitudinal displacement of steel rail
The detection speed and accuracy of detection.
In order to achieve the above objectives, the present invention adopts the following technical solutions:
One aspect of the present invention discloses a kind of longitudinal displacement of steel rail detection method based on image sequence, the method packet
It includes:
S1: coding maker is set on the fixed reference benchmark architecture and steel rail web beside rail;
S2: different time, different angle, the rail longitudinal direction position comprising all coding makers are acquired by image capture device
Phase shift transmission line live image forms image sequence;
S3: the marker detection method based on convolutional neural networks, the detection of building coding maker and location model, to image sequence
Coding maker in column is detected and is positioned;
S4: the characteristic point of each coding maker and the sub-pix image coordinate of characteristic point are decoded;
S5: the Three-dimension Reconstruction Model based on sub-pix image coordinate building longitudinal displacement of steel rail route scene;
S6: longitudinal displacement of steel rail is calculated based on the Three-dimension Reconstruction Model.
Preferably, the fixed reference benchmark architecture includes electrified column and/or bridge pull rod.
Preferably, the mode of the setting coding maker is spraying, listed or OEM.
Preferably, described image sequence includes the above longitudinal displacement of steel rail route live image of three width.
Preferably, the step S3 includes:
S31: a large amount of longitudinal displacement of steel rail route live images, longitudinal displacement of steel rail picture number of the building for training are acquired
According to collection;
S32: being trained study to the data set based on convolutional neural networks algorithm, generates coding maker detection and determines
Bit model;
S33: the coding maker in described image sequence is detected based on coding maker detection and location model
And positioning.
Preferably, the step S32 includes:
S321: it based on longitudinal displacement of steel rail route live image data set to be trained, obtains to be encoded in training image
The labeled data of mark;
S322: by longitudinal displacement of steel rail route live image data set and the labeled data input convolutional neural networks into
Row training generates coding maker detection and location model.
Preferably, step S5 includes:
S51: camera motion parameter matrix is estimated according to the sub-pix image coordinate of the characteristic point;
S52: empty using triangulation estimation characteristic point based on the corresponding camera motion parameter matrix of every width live image
Between coordinate;
S53: characteristic point space coordinate and camera motion matrix are optimized by light-stream adjustment, obtain live three-dimensional reconstruction knot
Fruit.
Preferably, step S6 includes:
S61: will be under the Three-dimension Reconstruction Model unification to the same coordinate system of each detection time point;
S62: the longitudinal displacement of steel rail between different detection time points is calculated based on the Three-dimension Reconstruction Model under the same coordinate system.
Another aspect of the present invention also discloses a kind of longitudinal displacement of steel rail detection system based on image sequence, the system
System includes:
Coding maker positioning unit, for based on being arranged on the fixed reference benchmark architecture and steel rail web beside rail
Coding maker, pass through image capture device acquire different time, different angle, the rail longitudinal direction position comprising all coding makers
Phase shift transmission line live image forms image sequence;
Positioning feature point unit, for the marker detection method based on convolutional neural networks, construct coding maker detection and
Location model is detected and is positioned to the coding maker in image sequence, and the characteristic point and feature of each coding maker are decoded
The sub-pix image coordinate of point;
Three-dimensional reconstruction unit, for the sub-pix image coordinate building longitudinal displacement of steel rail route scene based on characteristic point
Three-dimension Reconstruction Model;
It is displaced computing unit, for calculating longitudinal displacement of steel rail based on the Three-dimension Reconstruction Model.
Beneficial effects of the present invention are as follows:
The multi-angle of view longitudinal displacement of steel rail route scene that technical solution of the present invention is acquired based on different detection time points
Image sequence can accurately and efficiently detect longitudinal displacement of steel rail.The technology is to ensure the modernization operation peace of railway
Quick, accurate, reliable theories technique support is provided entirely.
Detailed description of the invention
Specific embodiments of the present invention will be described in further detail with reference to the accompanying drawing.
Fig. 1 shows a kind of flow chart of the longitudinal displacement of steel rail detection method based on image sequence of the present invention.
Fig. 2 shows showing for coding maker is arranged in a kind of longitudinal displacement of steel rail detection method based on image sequence of the present invention
Field schematic diagram.
Fig. 3 shows adoptable coding maker in a kind of longitudinal displacement of steel rail detection method based on image sequence of the present invention
Example.
Fig. 4 shows longitudinal displacement of steel rail route in a kind of longitudinal displacement of steel rail detection method based on image sequence of the present invention
The on-the-spot schematic of live image acquisition.
Fig. 5 shows the characteristic point of coding maker in a kind of longitudinal displacement of steel rail detection method based on image sequence of the present invention
Three-dimensional reconstruction schematic diagram.
Fig. 6 shows different detection time points three in a kind of longitudinal displacement of steel rail detection method based on image sequence of the present invention
Tie up reconstruction model diagram.
Fig. 7 shows different detection time points three in a kind of longitudinal displacement of steel rail detection method based on image sequence of the present invention
Dimension reconstruction model is transferred to the model schematic under the same coordinate system.
Specific embodiment
In order to illustrate more clearly of the present invention, the present invention is done further below with reference to preferred embodiments and drawings
It is bright.Similar component is indicated in attached drawing with identical appended drawing reference.It will be appreciated by those skilled in the art that institute is specific below
The content of description is illustrative and be not restrictive, and should not be limited the scope of the invention with this.
As shown in Figure 1, the invention discloses a kind of longitudinal displacement of steel rail detection method based on image sequence, the method
Include:
S1: coding maker is set on the fixed reference benchmark architecture and steel rail web beside rail.It can be in electrification
On the tracksides fixed reference benchmark architecture such as column, bridge pull rod and steel rail web, by spraying, being listed, the modes such as OEM are arranged
Coding maker, as shown in Figure 2.The various coding makers in photogrammetric can be used in coding maker, as shown in Figure 3.
S2: different time, different angle, the rail longitudinal direction position comprising all coding makers are acquired by image capture device
Phase shift transmission line live image forms image sequence.It is longitudinal rail can be acquired in each detection time point by image capture devices such as cameras
It is displaced scene image sequence, each image sequence includes the high-quality high-definition image of at least 3 width or more, different perspectives, each image
Comprising pre-arranged code marks all in scene, as shown in Figure 4.The longitudinal displacement of steel rail route live image of acquisition can store
It stores in equipment, or is uploaded on server, transfer use to follow-up equipment.
S3: image capture device acquisition image due to being affected by complicated factors such as weather, illumination, noises, and
Detecting target, proportion is smaller in the picture and change in location is larger, and conventional target localization method can not be accurate to coded markings
Positioning.Therefore, the present invention uses the marker detection method based on convolutional neural networks, constructs coding maker zone location model,
Coding maker in image sequence is detected and positioned.Wherein, the Practice Platform of convolutional neural networks uses Caffe,
Caffe is the deep learning frame based on C++, supports order line, Python and MATLAB interface, specific implementation method is such as
Under:
S31: the longitudinal displacement of steel rail route live image based on acquisition, longitudinal displacement of steel rail image of the building for training
Data set.Data images derive from a large amount of longitudinal displacement of steel rail route live images of testing staff's shooting.
S32: being trained study to data set based on convolutional neural networks method, generate coded markings region detection and
Location model.Based on longitudinal displacement of steel rail route live image data set to be trained, obtain to the coding mark in training image
Then longitudinal displacement of steel rail route live image data set and the labeled data are inputted convolutional Neural net by the labeled data of will
Network is trained, and generates coding maker detection and location model.According to the data set of above-mentioned building in the present invention, it is based on Caffe
Platform is trained study to data set using convolutional neural networks model, generates for detecting and location coding label
CaffeModel file.Wherein, CaffeModel file is the exclusive file generated based on Caffe platform training, supports order
Row, Python and MATLAB interface, can easily be called, output detection and positioning result.
S33: the coding maker in described image sequence is detected based on coding maker detection and location model
And positioning.
S4: the characteristic point of each coding maker and the sub-pix image coordinate of characteristic point are decoded.It can be several based on coding maker
The decoding of the information such as what structure, color, obtains the encoded radio and sub-pix exact image coordinate of characteristic point, as shown in Figure 5.
S5: the Three-dimension Reconstruction Model based on sub-pix image coordinate building longitudinal displacement of steel rail route scene, such as Fig. 6 institute
Show.Step S5 can include:
S51: camera motion parameter matrix is estimated according to the sub-pix image coordinate of the characteristic point;
S52: empty using triangulation estimation characteristic point based on the corresponding camera motion parameter matrix of every width live image
Between coordinate;
S53: characteristic point space coordinate and camera motion matrix are optimized by light-stream adjustment, obtain live three-dimensional reconstruction knot
Fruit.
S6: the longitudinal displacement of steel rail between different detection time points is calculated based on the Three-dimension Reconstruction Model.Utilize each detection
The characteristic point sub-pix image coordinate of coding maker in the image sequence of time point acquisition on trackside fixed reference benchmark architecture
Calculate the visual angle effect relationship of each image sequence.By under the Three-dimension Reconstruction Model unification to the same coordinate system of S5, different inspections are utilized
The three-dimensional reconstruction result for surveying the characteristic point in the image sequence of time point acquisition on steel rail web on coding maker calculates different inspections
The longitudinal displacement of steel rail between time point is surveyed, as shown in Figure 7.The longitudinal displacement of steel rail testing result of acquisition can store to be set in storage
It in standby, or is uploaded on server, transfers use to follow-up equipment.
The longitudinal displacement of steel rail detection system based on image sequence that the present invention discloses a kind of, the system comprises:
Coding maker positioning unit, for based on being arranged on the fixed reference benchmark architecture and steel rail web beside rail
Coding maker, pass through image capture device acquire different time, different angle, the rail longitudinal direction position comprising all coding makers
Phase shift transmission line live image forms image sequence.
Positioning feature point unit, for the marker detection method based on convolutional neural networks, construct coding maker detection and
Location model is detected and is positioned to the coding maker in image sequence, and the characteristic point and feature of each coding maker are decoded
The sub-pix image coordinate of point.
Three-dimensional reconstruction unit estimates camera motion parameter matrix and estimation for the sub-pix image coordinate based on characteristic point
Each characteristic point space coordinate constructs live Three-dimension Reconstruction Model.
It is displaced computing unit, it is longitudinal for calculating the rail between different detection time points based on the Three-dimension Reconstruction Model
Displacement.
The present invention is based on the coding makers being arranged on fixed reference benchmark architecture and steel rail web, acquire image sequence
Then column use the method based on deep learning, are detected and be accurately positioned to coded markings, overcome picture noise and non-phase
Close information interference.It is guidance with deep learning positioning result, decodes and obtain based on information such as coded markings geometry, colors
The encoded radio and sub-pix image coordinate of characteristic point.According to the characteristic point sub-pix image coordinate in each image sequence, to every
A longitudinal displacement of steel rail scene carries out three-dimensional reconstruction.Recycle trackside fixed reference in the image sequence of each detection time point acquisition
The characteristic point sub-pix image coordinate of coding maker on benchmark architecture, calculates the visual angle effect relationship of each image sequence, by three
It ties up under reconstructed results unification to the same coordinate system, is encoded on steel rail web in the image sequence acquired using different detection time points
The three-dimensional reconstruction result of characteristic point on mark can finally calculate the longitudinal displacement of steel rail between different detection time points.
The present invention is suitable for the non-contact accurate detection of longitudinal displacement of steel rail, is not necessarily to testing staff's upper track, without in route
Upper setting caliberating device sets fixed imaging device without trackside, is not necessarily to secondary light source, coding maker setup cost is extremely low, right
Railroad embankment is had no effect, detect simple and quick and detection accuracy it is high, can further genralrlization be applied to such as road frog rail
In the displacement detecting of other critical structural components, to ensure that the operation security of railway provides quick, accurate, reliable theory and technology
It supports.
Obviously, the above embodiment of the present invention be only to clearly illustrate example of the present invention, and not be pair
The restriction of embodiments of the present invention may be used also on the basis of the above description for those of ordinary skill in the art
To make other variations or changes in different ways, all embodiments can not be exhaustive here, it is all to belong to this hair
The obvious changes or variations that bright technical solution is extended out are still in the scope of protection of the present invention.
Claims (9)
1. a kind of longitudinal displacement of steel rail detection method based on image sequence, which is characterized in that the described method includes:
S1: coding maker is set on the fixed reference benchmark architecture and steel rail web beside rail;
S2: different time is acquired by image capture device, different angle, includes the longitudinal displacement of steel rail line of all coding makers
Road live image forms image sequence;
S3: the marker detection method based on convolutional neural networks, the detection of building coding maker and location model, in image sequence
Coding maker detected and positioned;
S4: the characteristic point of each coding maker and the sub-pix image coordinate of characteristic point are decoded;
S5: the Three-dimension Reconstruction Model based on sub-pix image coordinate building longitudinal displacement of steel rail route scene;
S6: longitudinal displacement of steel rail is calculated based on the Three-dimension Reconstruction Model.
2. the method according to claim 1, wherein the fixed reference benchmark architecture includes electrified column
And/or bridge pull rod.
3. the method according to claim 1, wherein the mode of the setting coding maker be spraying, it is listed or
OEM.
4. the method according to claim 1, wherein described image sequence includes the above longitudinal displacement of steel rail of three width
Route live image.
5. the method according to claim 1, wherein the step S3 includes:
S31: a large amount of longitudinal displacement of steel rail route live images, longitudinal displacement of steel rail image data of the building for training are acquired
Collection;
S32: study is trained to the data set based on convolutional neural networks algorithm, generates coding maker detection and positioning mould
Type;
S33: the coding maker in described image sequence is detected and is determined based on coding maker detection and location model
Position.
6. according to the method described in claim 5, it is characterized in that, the step S32 includes:
S321: it based on longitudinal displacement of steel rail route live image data set to be trained, obtains to coding maker in training image
Labeled data;
S322: longitudinal displacement of steel rail route live image data set and labeled data input convolutional neural networks are instructed
Practice, generates coding maker detection and location model.
7. the method according to claim 1, wherein step S5 includes:
S51: camera motion parameter matrix is estimated according to the sub-pix image coordinate of the characteristic point;
S52: it is sat based on the corresponding camera motion parameter matrix of every width live image using the triangulation estimation feature space of points
Mark;
S53: characteristic point space coordinate and camera motion matrix are optimized by light-stream adjustment, obtain live three-dimensional reconstruction result.
8. the method according to claim 1, wherein step S6 includes:
S61: will be under the Three-dimension Reconstruction Model unification to the same coordinate system of each detection time point;
S62: the longitudinal displacement of steel rail between different detection time points is calculated based on the Three-dimension Reconstruction Model under the same coordinate system.
9. a kind of longitudinal displacement of steel rail detection system based on image sequence, which is characterized in that the system comprises:
Coding maker positioning unit, for based on the volume being arranged on the fixed reference benchmark architecture and steel rail web beside rail
Code mark acquires different time by image capture device, different angle, includes the longitudinal displacement of steel rail line of all coding makers
Road live image forms image sequence;
Positioning feature point unit constructs coding maker detection and positioning for the marker detection method based on convolutional neural networks
Model detected and positioned to the coding maker in image sequence, and the characteristic point and characteristic point of each coding maker are decoded
Sub-pix image coordinate;
Three-dimensional reconstruction unit, the three-dimensional for the sub-pix image coordinate building longitudinal displacement of steel rail route scene based on characteristic point
Reconstruction model;
It is displaced computing unit, for calculating longitudinal displacement of steel rail based on the Three-dimension Reconstruction Model.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710388790.9A CN107330931B (en) | 2017-05-27 | 2017-05-27 | A kind of longitudinal displacement of steel rail detection method and system based on image sequence |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710388790.9A CN107330931B (en) | 2017-05-27 | 2017-05-27 | A kind of longitudinal displacement of steel rail detection method and system based on image sequence |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107330931A CN107330931A (en) | 2017-11-07 |
CN107330931B true CN107330931B (en) | 2019-10-25 |
Family
ID=60192948
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710388790.9A Active CN107330931B (en) | 2017-05-27 | 2017-05-27 | A kind of longitudinal displacement of steel rail detection method and system based on image sequence |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107330931B (en) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103821054A (en) * | 2014-03-12 | 2014-05-28 | 武汉大学 | INS (inertial navigation system) and total station combination-based track geometrical state measurement system and method |
CN106204620A (en) * | 2016-07-21 | 2016-12-07 | 清华大学 | A kind of tactile three-dimensional power detection method based on micro-vision |
CN106199570A (en) * | 2016-07-20 | 2016-12-07 | 上海自仪泰雷兹交通自动化系统有限公司 | A kind of track train displacement and speed detection system |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP3009984A1 (en) * | 2014-10-14 | 2016-04-20 | Sick Ag | Detection system for optical codes |
US9943247B2 (en) * | 2015-07-28 | 2018-04-17 | The University Of Hawai'i | Systems, devices, and methods for detecting false movements for motion correction during a medical imaging scan |
-
2017
- 2017-05-27 CN CN201710388790.9A patent/CN107330931B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103821054A (en) * | 2014-03-12 | 2014-05-28 | 武汉大学 | INS (inertial navigation system) and total station combination-based track geometrical state measurement system and method |
CN106199570A (en) * | 2016-07-20 | 2016-12-07 | 上海自仪泰雷兹交通自动化系统有限公司 | A kind of track train displacement and speed detection system |
CN106204620A (en) * | 2016-07-21 | 2016-12-07 | 清华大学 | A kind of tactile three-dimensional power detection method based on micro-vision |
Also Published As
Publication number | Publication date |
---|---|
CN107330931A (en) | 2017-11-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109167956B (en) | Full-bridge surface moving load spatial distribution monitoring system | |
CN106407315B (en) | A kind of vehicle autonomic positioning method based on street view image database | |
CN103985254B (en) | A kind of multi-view point video for large scene traffic monitoring merges and traffic parameter acquisition method | |
CN103425967B (en) | A kind of based on stream of people's monitoring method of pedestrian detection and tracking | |
CN104575003B (en) | A kind of vehicle speed detection method based on traffic surveillance videos | |
CA2839984C (en) | Device for measuring speed and position of a vehicle moving along a guidance track, method and computer program product corresponding thereto | |
CN103150559B (en) | Head recognition and tracking method based on Kinect three-dimensional depth image | |
CN101847206B (en) | Pedestrian traffic statistical method and system based on traffic monitoring facilities | |
CN103177247B (en) | A kind of object detection method merging various visual angles information | |
CN108375377B (en) | Device and method for determining the position of a vehicle on a track | |
CN106960179B (en) | Rail line Environmental security intelligent monitoring method and device | |
CN104748685A (en) | Dynamic measurement method of geometric parameters of overhead contact system | |
CN106871805A (en) | vehicle-mounted rail gauge measuring system and measuring method | |
JP2020002774A (en) | Leading rail vehicle and method with remote controlled vehicle | |
CN107851125A (en) | The processing of two step object datas is carried out by vehicle and server database to generate, update and transmit the system and method in accurate road characteristic data storehouse | |
CN104634222A (en) | Distance-measuring system and distance-measuring method | |
CN103456172A (en) | Traffic parameter measuring method based on videos | |
CN105260719A (en) | Railway platform line-crossing detection method | |
CN106199570A (en) | A kind of track train displacement and speed detection system | |
CN103729860B (en) | A kind of method and apparatus of tracking image target | |
CN204514232U (en) | A kind of range measurement system | |
CN104973092A (en) | Rail roadbed settlement measurement method based on mileage and image measurement | |
CN109360225A (en) | A kind of optimization system and method for motion model | |
CN108986150A (en) | A kind of image light stream estimation method and system based on non-rigid dense matching | |
CN106295491A (en) | Track line detection method and device |
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 |