CN109870683A - Rail exception fastener detection method based on radar signal Time-frequency Analysis - Google Patents
Rail exception fastener detection method based on radar signal Time-frequency Analysis Download PDFInfo
- Publication number
- CN109870683A CN109870683A CN201910168959.9A CN201910168959A CN109870683A CN 109870683 A CN109870683 A CN 109870683A CN 201910168959 A CN201910168959 A CN 201910168959A CN 109870683 A CN109870683 A CN 109870683A
- Authority
- CN
- China
- Prior art keywords
- time
- fastener
- frequency spectrum
- signal
- frequency
- 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
Links
Landscapes
- Train Traffic Observation, Control, And Security (AREA)
Abstract
The rail exception fastener detection method based on radar signal Time-frequency Analysis that the present invention provides a kind of is overhauled for rail.This method acquires data by the unmanned plane equipped with 24GHz continuous wave radar, and unmanned plane emits 24GHz continuous wave signal when reaching the upper empty position of railway rail clip along rail inspection, acquires before train process, the echo data in and after warp.This method carries out radar monitoring to normal fastener first, echo-signal is subjected to Short Time Fourier Transform, the time-frequency spectrum classification deposit database of acquisition, then respective classes time-frequency spectrum in the time-frequency spectrum and database of fastener to be detected is compared one by one, judges whether railway rail clip is abnormal by the otherness of time-frequency spectrum.Operation of the present invention is simple, and detection efficiency and accuracy rate are high, can efficiently and accurately judge whether railway rail clip exception occurs.
Description
Technical field
The present invention relates to railway maintenance and Radar Signal Processing Technology field, more particularly to it is a kind of when being based on radar signal
The abnormal fastener detection method of frequency signature analysis.
Background technique
In recent years, railway is fast-developing, and bullet train brings more convenience to people's trip, improves the trip of the people
Efficiency.Railway construction is even more that outstanding effect is played in economic construction of China, it has furthered the distance between city and city,
It solves the problems, such as transport capacity deficiency, is to push the important motivity of economic sustained and rapid development, therefore the development of railway is to state
Family has highly important strategic position.We should more pay close attention to railway operation peace while enjoying railway bring benefit
Full problem.In addition to train itself, the safety problem of rail track also be can not be ignored.During the inspection of route, whether rail
Intact, whether having foreign matter etc. on route is all necessary detection project.Wherein, fastener is the centre part for being coupled rail and sleeper,
As the critical component of fixed rail, guarantee that it is in normal condition, there is vital meaning in guaranteeing line security.
Currently, China is mainly using the mode of manual inspection, time-consuming and laborious by visually being detected, omission factor is higher.For
The automatic measurement technique of abnormal fastener has optical image security technology, detect although this technology is capable of greater efficiency,
But it is susceptible to the influence of resolution ratio, light and environmental factor, thus there is certain limitation.How to abnormal fastener into
Row efficient detection is a urgent problem to be solved.
Summary of the invention
Aiming at the problems existing in the prior art, the present invention provides a kind of rail based on radar signal Time-frequency Analysis
Abnormal fastener detection method, to solve the automatic test problems of rail exception fastener.
Rail exception fastener detection method provided by the invention based on radar signal Time-frequency Analysis, the hardware of realization
Including unmanned plane and remote computer；Equipped with 24GHz continuous wave radar and data acquisition module on the unmanned plane；It is described
Unmanned plane along rail inspection, 24GHz continuous wave signal is emitted when reaching the upper empty position of railway rail clip, the data are adopted
Collect module and acquires radar return data；Described is that database and data processing module are provided on computer.Detection of the invention
Method realizes that step includes the following steps 1~step 3.
Step 1, normal fastener database is established, comprising: radar is directed at normal fastener and emits 24GHz continuous wave signal, adopts
Collect in the case that train process before, pass through when and warp after three kinds normal fastener echo-signal.It will be collected normal
The echo-signal of fastener carries out Short Time Fourier Transform, obtains the two-dimentional time-frequency spectrum of signal, and time-frequency spectrum is stored in data
Library；Database pass through according to train before, time-frequency spectrum is divided into three classes storage by three kinds of situations when passing through and after warp, under every class
Subclass is set also according to the frequency range of signal, each subclass represents a frequency range；This step deposits time-frequency spectrum
Enter under corresponding situation classification in the subclass of respective frequencies range.
Step 2, open detection mode, radar contacting piece emit 24GHz continuous wave signal, and the echo-signal of acquisition is carried out
Short Time Fourier Transform obtains the two-dimentional time-frequency spectrum X of signal, and time-frequency spectrum X and corresponding acquisition situation are passed at data
Manage module；Acquisition situation refers to before train process, acquires when passing through or after warp.
Step 3, abnormal fastener detection.Data processing module is by incoming time-frequency spectrum X and situation class corresponding in database
Time-frequency spectrum in the subclass of respective frequencies range under not is compared one by one, calculates the time-frequency under X and corresponding subclass
The coefficient of variation of spectrogram；When coefficient of variation is greater than preset threshold, it is determined as abnormal fastener, otherwise, it is determined that be normal fastener, and
It will be in the respective classes of the time-frequency spectrum deposit database of fastener.
The coefficient of variation is defined as:
Compared with the existing technology, the advantages and positive effects of the present invention are: (1) present invention passed through according to train before, in
With rear three kinds of situations, detections of radar is carried out to normal fastener, obtains corresponding time-frequency spectrum, the data acquisition of three kinds of situations makes
The detection for obtaining contacting piece is more accurate, comprehensive；(2) since the speed of train is different, the frequency of echo-signal also will be different, therefore
The method of the present invention divides subclass according to the frequency range of weak signal, and the database being built such that, data acquisition pattern is simple,
The acquisition of normal data is also easier；And when detection, it is only necessary to according to the frequency range of the weak signal of fastener to be measured,
Subclass in subclass corresponding to database is compared, so that detection is more rapid, data calculation amount is few；(3) this hair
Bright method it is easy to operate, have round-the-clock, detection efficiency and high accuracy for examination, can efficiently and accurately judge railway rail clip
The problems such as whether there is exception, for example loosening.
Detailed description of the invention
Fig. 1 is the normal fastener schematic diagram of rail；
Fig. 2 is rail exception fastener schematic diagram；
Fig. 3 is the flow diagram of rail exception fastener detection method provided by the invention；
Fig. 4 is train by preceding normal fastener signal time-frequency spectrum；
Fig. 5 is normal fastener signal time-frequency spectrum when train passes through；
Fig. 6 is train warp normal fastener signal time-frequency spectrum later；
Fig. 7 is signal time-frequency spectrum when people pulls abnormal fastener；
Fig. 8 is exception fastener signal time-frequency spectrum when train passes through.
Specific embodiment
In order to keep the purposes, technical schemes and advantages of the embodiment of the present invention clearer, below in conjunction with attached drawing and implementation
The present invention is described in further detail for example.
Railway rail clip is on track to the important part for being coupled rail and sleeper, also known as intermediate fastening.It is acted on
It is to guarantee being reliably connected between rail and sleeper long-term effectively, prevents the longitudinal movement between rail and sleeper, it is ensured that rail
Absorbing shock performance is given full play to away from normal, and under the effect of the power of train, slows down the accumulation of line residual deformation.Rail is just
Normal fastener is as shown in Figure 1, with the vibration of rail microvibration can occur for the normal fastener of rail when train passes through.When rail button
Part is when in an abnormal state, such as when fastener loosening, as shown in Fig. 2, during train passes through, in addition to following rail to shake
It is dynamic, vibration by a larger margin itself can be generated, vibration can cause the radial velocity of opposite radar, and here it is the present invention is based on radars
The basis of signal detection exception fastener.
The hardware that the present invention realizes includes unmanned plane and remote computer.There is 24GHz continuous wave radar in UAV flight
And data acquisition module.Database and data processing module are built on the remote computer.Rail provided by the invention is detained extremely
The process of part detection method, as shown in figure 3, overall work mode is divided into training mode and detection pattern.According to different work
The two-dimentional time-frequency spectrum of echo-signal is passed to different modules by mode.When being in training mode, two-dimentional time-frequency spectrum is passed
Enter database, the time-frequency spectrum of normal fastener is only stored in database；When be in detection pattern when, by collected fastener when
Spectrogram is passed to data processing module.Normal fastener database is established in training mode, and contacting piece carries out in a detection mode
Real-time detection.
Firstly, illustrating the realization of data acquisition of the invention.Unmanned plane follows train along rail inspection, when unmanned plane reaches
The fastener of radar downwards emits 24GHz continuous wave signal when the upper empty position of railway rail clip, acquires echo signal data.Nobody
The position of machine acquires data in the oblique upper of fastener under three circumstances: passing through before train process, when train passes through with train
Afterwards.Unmanned plane generally stops 1~15 second time in the oblique overhead of fastener.
Secondly, the basis of detections of radar Vibration Targets of the present invention is Doppler effect.Based on Doppler effect, when object phase
When having radial velocity to radar, reflection signal can generate Doppler frequency shift, and frequency displacement and speed have following relationship:
Wherein, λ is the wavelength of radar, and v is the radial velocity of target, and f is the Doppler frequency shift that target generates.Abnormal fastener
The radial velocity of opposite radar can be generated in vibration, thus by the Doppler frequency shift of measurement echo-signal, it can be realized different
The detection of normal fastener.Since random vibration can occur for abnormal fastener, thus the radial velocity of opposite radar is time-varying, how general
Strangling frequency displacement is also time-varying.
In order to show the vibrational state of abnormal fastener, contacting piece echo-signal of the present invention carries out time frequency analysis.When common
Frequency analysis method includes Short Time Fourier Transform, Wigner-Ville distribution and wavelet transformation etc..Short Time Fourier Transform has
Calculate simple, the advantages that cross term is not present, thus the present invention using Short Time Fourier Transform show radar echo signal when
Spectral characteristic.The formula of Short Time Fourier Transform can be expressed as:
Wherein, s (τ) indicates that radar echo signal, w (τ-t) indicate that window function, subscript * indicate conjugation, and t indicates time, f
Indicate the frequency of echo-signal, τ indicates time delay, and j is imaginary unit.
The resolving power of Short Time Fourier Transform depends entirely on the when window radius Δ of window functiontWith frequency window radius Δw.But
Window function when window radius and the value of frequency window radius be not independent from each other, according to uncertainty principle, there is following relationship:
As can be seen that time sense and frequency resolution are always conflicting for Short Time Fourier Transform.?
In the time-frequency spectrum of signal, high time sense and frequency resolution cannot be obtained simultaneously, one of them narrows, another just must
Surely it broadens.This, which means that, can only sacrifice time sense to exchange higher frequency resolution for, or in turn with frequency point
The reduction of power is distinguished to exchange the raising of time sense for.When window function is Gaussian function, equal sign is set up, time and frequency discrimination
Power compromise.Therefore, window function of the invention uses Gauss function.
The realization step for illustrating rail exception fastener detection method of the invention below, as shown in figure 3, including the following steps
1 to step 3.
Step 1, in training mode, normal fastener database is established.
Unmanned plane follows train along rail inspection, respectively in the case that train process before, pass through when and warp after three kinds,
Radar carries out the transmitting of 24GHz continuous wave signal to normal fastener, acquires the echo-signal of normal fastener in these three cases.
The echo-signal of collected normal fastener is carried out Short Time Fourier Transform by data acquisition module, obtains the two dimension of echo-signal
Time-frequency spectrum, and time-frequency spectrum is passed to database purchase.In the training stage, data collecting module collected is normal button
The echo-signal of part.Database pass through according to train before, time-frequency spectrum is divided into three classes and deposits by three kinds of situations when passing through and after warp
Storage.
Fig. 4, Fig. 5 and Fig. 6 are radar return letter of the normal fastener before train process, when passing through and after warp respectively
Number.Train in the present invention is by preceding and refer to that after, train will be by before fastener and train will be after fastener.Fig. 4
First half be train pass through before, latter half be train passing through；The first half of Fig. 6 is that train is passing through, after
Half part is that train has been moved off, so being that there are train bodies and two section vehicles in the latter half of Fig. 4 and the first half of Fig. 6
The reflecting component in compartment gap.It can be found that when in a bit of time after being passed through by preceding and train that runs a train, rail, stone
And fastener etc. can still vibrate.When train passes through, the instantaneous frequency of echo-signal can be divided into two parts, a part
It is strong echo-signal, a part is weak echo signal.Strong echo-signal is to be reflected to generate by train main body.When train passes through
When, it has car body and is shaken with respect to the car body generated in the velocity component and train traveling process of radar, to generate Doppler
Frequency displacement.Since, there are gap, periodic regularity is presented in the reflecting component of train body between two section compartments.Weak echo signal
It is by generations such as rail, stone, fasteners, the vibration frequency of these reverberations is significantly greater than the vibration of train body generation.But
It is the signal that frequency intensity is less than train body reflection.At this point, fastener is securely held on rail, thus phase is generated with rail
Same vibration.
Strong echo-signal and the frequency range of weak echo signal are all the speed with train and change.Therefore present invention side
Multiple subclass are arranged under three circumstances, also according to the different frequency scope of signal in method.Each subclass represents a frequency
Range, the embodiment of the present invention are that multiple subclass are segmented according to the frequency range of weak echo signal.For example, after to train warp
In the case of echo data, the frequency range of the weak echo signal in the time-frequency map of acquisition is 0-1000Hz, then passes through in train
Later it under classification, by the subclass of map deposit 0-1000Hz, if there is no the subclass of such frequency range, then builds
The subclass of vertical 0-1000Hz.In Fig. 5, the frequency range for collecting weak echo signal is 500-1500Hz, then establishes 500-
The subclass of 1500Hz.The frequency range of different subclass is can be partly overlapping.
During real-time detection, the data of fastener to be measured are according to three in the frequency range and database of weak echo signal
The data of corresponding subclass in major class compare one by one, to be detected.
Step 2, in a detection mode, unmanned plane follows train along rail inspection, when reaching empty position on fastener, radar
Contacting piece emits 24GHz continuous wave signal, and the echo-signal of acquisition is carried out Short Time Fourier Transform by data acquisition module, obtains
The two-dimentional time-frequency spectrum of signal, is passed to data processing module for time-frequency spectrum.
Step 3, data processing module will be in the subclass that corresponded under situation classification in incoming time-frequency spectrum and database
Time-frequency spectrum be compared one by one, calculate X and the time-frequency spectrum in corresponding subclass coefficient of variation, when coefficient of variation is greater than
When preset threshold, it is determined as abnormal fastener, otherwise, it is determined that being normal fastener.When being normal fastener, by the time-frequency spectrum of fastener
It is stored in the subclass of corresponding situation, updates database data.
When fastener loosens, for fastener in addition to following rail to vibrate, itself can generate vibration by a larger margin.Fig. 7 is
People pulls the echo-signal time-frequency spectrum of abnormal fastener.With the pulling of people, abnormal fastener can generate irregular vibration, due to
The frequency that people pulls is lower, and the highest frequency of fastener vibration can reach 100Hz.When train presses through, the vibration frequency of abnormal fastener
Rate is higher, and the frequency intensity that fastener generates is lower than the frequency intensity that train body generates.
Fig. 8 is the time-frequency spectrum of exception fastener when train passes through.For security reasons, what the train that the present invention acquires passed through
Fastener is normal fastener, so the echo-signal of the exception fastener is reasonable artificial synthesized signal.Due to abnormal fastener
Vibration is to be pulled to generate by people, so fastener vibration frequency is lower, highest frequency 100Hz is overlapped with train body frequency.
When train is by abnormal fastener, fastener meeting sustained vibration can also be vibrated when two section compartment junctions are passed through.Therefore,
Comparison diagram 5 and Fig. 8, in two section compartment gaps, the frequency component intensity that train generates is smaller, and the frequency that abnormal fastener generates
Component intensity is stronger, therefore there are notable differences for two time-frequency spectrums, that is, can determine that the fastener for abnormal fastener.When train passes through
When abnormal fastener, the vibration frequency of fastener is higher, so the time-frequency spectrum difference of normal fastener and abnormal fastener is more obvious, inspection
It is higher to survey accuracy.
When being in detection pattern, collected fastener echo-signal time-frequency spectrum is passed into data processing module.It will be new
The signal time-frequency spectrum of acquisition compares one by one with the time-frequency spectrum of respective classes in database, calculates the time-frequency spectrum and database
The coefficient of variation of middle time-frequency spectrum.Coefficient of variation is defined as in database respective classes having with fastener time-frequency spectrum to be measured obvious
The ratio of category time-frequency spectrum total quantity in the quantity and database of the time-frequency spectrum of difference, it may be assumed that
The difference of respective classes time-frequency spectrum is smaller in the time-frequency spectrum and database of normal fastener, thus coefficient of variation also compared with
It is small；Abnormal fastener can also generate vibration by a larger margin other than following rail vibration, thus coefficient of variation is larger.If difference
Coefficient is greater than threshold value, then determines the fastener for abnormal fastener；If coefficient of variation is less than threshold value, determine that the fastener is normal button
Part, and the time-frequency spectrum is passed to database, update database data.Threshold value can be set to the arbitrary value in 0 to 1, this
It is 0.5 that threshold value is selected in inventive embodiments, i.e., in the time-frequency spectrum and database of fastener to be measured in respective classes time-frequency spectrum
More than half there are when notable difference, which is judged as abnormal fastener.Threshold value is arranged lower, and the sensitivity of detection is got over
It is high.
When two two-dimentional time-frequency spectrums are compared, by the two-dimentional time-frequency spectrum of fastener to be measured and database when
The pixel value of spectrogram subtracts each other, and obtains the difference value of the two, if difference value is greater than α times of total pixel value, then it is assumed that fastener to be measured
Two-dimentional time-frequency spectrum and database respective classes time-frequency spectrum difference it is obvious.Wherein, α is threshold value, value interval be (0,
1)。
Obviously, described embodiment is also only a part of the embodiments of the present invention, rather than whole embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
Claims (6)
1. a kind of rail exception fastener detection method based on radar signal Time-frequency Analysis, which is characterized in that realization it is hard
Part includes unmanned plane and remote computer；Equipped with 24GHz continuous wave radar and data acquisition module on the unmanned plane；Institute
The unmanned plane stated emits 24GHz continuous wave signal to fastener when reaching the upper empty position of railway rail clip along rail inspection, described
Data collecting module collected echo data；Database and data processing module are provided on the computer；The inspection
Survey method detects abnormal fastener, is included the following steps: by the Doppler frequency shift of measurement echo-signal
Step 1, normal fastener database is established, comprising: radar emits 24GHz continuous wave signal, acquisition to normal railway rail clip
In the case that train process before, pass through when and warp after three kinds normal fastener echo-signal；By the radar return of acquisition
Signal carries out Short Time Fourier Transform, obtains the two-dimentional time-frequency spectrum of signal, time-frequency spectrum is stored in database；Database according to
Train pass through before, time-frequency spectrum is divided into three classes storage by three kinds of situations when passing through and after warp, also according to signal under every class
Subclass is arranged in frequency range, and each subclass represents a frequency range；Time-frequency spectrum is stored in corresponding situation class by this step
It does not descend in the subclass of respective frequencies range；
Step 2, open detection mode, radar contacting piece emit 24GHz continuous wave signal, and the echo-signal of acquisition is carried out in short-term
Fourier transformation obtains the two-dimentional time-frequency spectrum X of signal, and time-frequency spectrum X and corresponding acquisition situation are passed to data processing mould
Block；Acquisition situation refers to before train process, acquires when passing through or after warp；
Step 3, data processing module is by incoming time-frequency spectrum X and the respective frequencies range under situation classification corresponding in database
Subclass in time-frequency spectrum be compared one by one, calculate the coefficient of variation of the time-frequency spectrum under X and corresponding subclass；It is on duty
When different coefficient is greater than preset threshold, it is determined as abnormal fastener, otherwise, it is determined that being normal fastener, and the time-frequency spectrum of fastener is deposited
Enter in the respective classes of database；
The coefficient of variation is defined as:
2. the method according to claim 1, wherein the unmanned plane stops 1 positioned at the oblique overhead of fastener
~15 seconds.
3. the method according to claim 1, wherein adopt when Short Time Fourier Transform in the step 1
Window function is Gauss function.
4. the method according to claim 1, wherein in the step 1, according under every class also according to weak time
Subclass is arranged in the frequency range of wave signal.
5. the method according to claim 1, wherein two time-frequency spectrums are compared in the step 3
When, the pixel value of the time-frequency spectrum in the time-frequency spectrum and database of fastener to be measured is subtracted each other, the difference value of the two is obtained, if poor
Total pixel value of the different value greater than time-frequency spectrum in α times of database, then it is assumed that the time-frequency spectrum of fastener to be measured and database when
Spectrogram difference is obvious, and α is setting threshold value, and value interval is (0,1).
6. method according to any one of claims 1 to 4, which is characterized in that preset threshold value is 0.5 in the step 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910168959.9A CN109870683B (en) | 2019-03-06 | 2019-03-06 | Rail abnormal fastener detection method based on radar signal time-frequency characteristic analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910168959.9A CN109870683B (en) | 2019-03-06 | 2019-03-06 | Rail abnormal fastener detection method based on radar signal time-frequency characteristic analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109870683A true CN109870683A (en) | 2019-06-11 |
| CN109870683B CN109870683B (en) | 2020-09-04 |
Family
ID=66919876
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910168959.9A Active CN109870683B (en) | 2019-03-06 | 2019-03-06 | Rail abnormal fastener detection method based on radar signal time-frequency characteristic analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109870683B (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000230208A (en) * | 1999-02-12 | 2000-08-22 | Kobe Steel Ltd | Slackness inspecting device for rail fastening device |
| JP2007147412A (en) * | 2005-11-27 | 2007-06-14 | Teruya:Kk | Real-time inspection system of loose rail fastening using dc battery-less rfid tag with sensor input functions |
| CN105501248A (en) * | 2016-02-16 | 2016-04-20 | 株洲时代电子技术有限公司 | Railway line inspection system |
| US20160236698A1 (en) * | 2015-02-16 | 2016-08-18 | Electro-Motive Diesel, Inc. | Automatic Disabling of Unpowered Locked Wheel Fault Detection for Slipped Traction Motor Pinion |
| CN106815552A (en) * | 2016-12-09 | 2017-06-09 | 云南航天工程物探检测股份有限公司 | Data signal post-processing approach based on time frequency analysis |
| CN107403139A (en) * | 2017-07-01 | 2017-11-28 | 南京理工大学 | A kind of municipal rail train wheel flat fault detection method |
| CN107748862A (en) * | 2017-09-21 | 2018-03-02 | 清华大学 | A kind of unmanned plane sorting technique and device based on dual-frequency radar signal time-frequency distributions |
| CN108058721A (en) * | 2018-01-26 | 2018-05-22 | 北京市劳动保护科学研究所 | A kind of rail fastener loose condition detection method and system |
| CN207946369U (en) * | 2017-10-13 | 2018-10-09 | 成都精工华耀科技有限公司 | A kind of railway rail clip abnormal detector based on pointolite array linear array imaging |
| CN109409225A (en) * | 2018-09-21 | 2019-03-01 | 清华大学 | Unmanned plane classification method and device based on the fusion of radar multipath signal time-frequency characteristics |
-
2019
- 2019-03-06 CN CN201910168959.9A patent/CN109870683B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000230208A (en) * | 1999-02-12 | 2000-08-22 | Kobe Steel Ltd | Slackness inspecting device for rail fastening device |
| JP2007147412A (en) * | 2005-11-27 | 2007-06-14 | Teruya:Kk | Real-time inspection system of loose rail fastening using dc battery-less rfid tag with sensor input functions |
| US20160236698A1 (en) * | 2015-02-16 | 2016-08-18 | Electro-Motive Diesel, Inc. | Automatic Disabling of Unpowered Locked Wheel Fault Detection for Slipped Traction Motor Pinion |
| CN105501248A (en) * | 2016-02-16 | 2016-04-20 | 株洲时代电子技术有限公司 | Railway line inspection system |
| CN106815552A (en) * | 2016-12-09 | 2017-06-09 | 云南航天工程物探检测股份有限公司 | Data signal post-processing approach based on time frequency analysis |
| CN107403139A (en) * | 2017-07-01 | 2017-11-28 | 南京理工大学 | A kind of municipal rail train wheel flat fault detection method |
| CN107748862A (en) * | 2017-09-21 | 2018-03-02 | 清华大学 | A kind of unmanned plane sorting technique and device based on dual-frequency radar signal time-frequency distributions |
| CN207946369U (en) * | 2017-10-13 | 2018-10-09 | 成都精工华耀科技有限公司 | A kind of railway rail clip abnormal detector based on pointolite array linear array imaging |
| CN108058721A (en) * | 2018-01-26 | 2018-05-22 | 北京市劳动保护科学研究所 | A kind of rail fastener loose condition detection method and system |
| CN109409225A (en) * | 2018-09-21 | 2019-03-01 | 清华大学 | Unmanned plane classification method and device based on the fusion of radar multipath signal time-frequency characteristics |
Non-Patent Citations (2)
| Title |
|---|
| YIQI XIA等: ""Broken railway fastener detection based on Adaboost algorithm"", 《2010 INTERNATIONAL CONFERENCE ON OPTOELECTRONICS AND IMAGE PROCESSING》 * |
| 赵珊珊等: ""基于SIFT特征的铁路扣件状态检测算法"", 《传感器与微系统》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109870683B (en) | 2020-09-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9341707B2 (en) | Method for classifying moving vehicles | |
| CN103593979B (en) | For by tracking the method that vehicle position parameter classifies driving vehicle | |
| JP6040096B2 (en) | Road surface condition estimation device | |
| CA2503581C (en) | Weather and airborne clutter suppression using a cluster shape classifier | |
| US9121934B2 (en) | Radar system and detection method | |
| US20080129582A1 (en) | Tracking target method of radar with frequency modulation continuous wave | |
| CN101876705A (en) | Frequency domain vehicle detecting method based on single-frequency continuous wave radar | |
| US20090303109A1 (en) | Method for filtering sea clutter in a radar echo using a hydrographic model | |
| CN107238821A (en) | The airfield runway foreign matter detecting method and device of a kind of feature based spectrum signature | |
| CN106199573A (en) | It is mounted laterally radar for what banister controlled | |
| CN105738898A (en) | Multilane radar speed measurement method and device based on combined distance, angle and speed measurement | |
| CN105989710B (en) | A kind of device for monitoring vehicle and method based on audio | |
| CN104407346A (en) | Mobile runway foreign object debris (FOD) monitoring method based on information integration | |
| CN104112305A (en) | Passenger flow monitoring method and apparatus based on dual laser ranging | |
| JPH08501877A (en) | Method and apparatus for determining relative visibility | |
| CN104199002A (en) | Radar information processing method used for restraining false-alarm objects of rotary blades of wind driven generators | |
| CN104007428B (en) | A kind of method identifying the non-meteorological echo that running car causes in radar return | |
| CN109870683A (en) | Rail exception fastener detection method based on radar signal Time-frequency Analysis | |
| CN113189590A (en) | FOD-based target identification method and device and storage medium | |
| CN111538002B (en) | FOD detection method based on vehicle-mounted millimeter wave radar | |
| Jang et al. | Evaluation of traffic data accuracy using Korea detector testbed | |
| US6505508B1 (en) | Apparatus for detecting turbulent layer | |
| RU2599870C1 (en) | Method of discrete noise classifying and blanking | |
| Santos et al. | Sensing techniques and detection methods for train approach detection | |
| CN103150905A (en) | Method for installing wave frequency detector at roadside to detect traffic flow |
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 |