CN102332089B - Railway wagon brake shoe key going-out fault recognition method based on artificial neural network - Google Patents
Railway wagon brake shoe key going-out fault recognition method based on artificial neural network Download PDFInfo
- Publication number
- CN102332089B CN102332089B CN 201110172116 CN201110172116A CN102332089B CN 102332089 B CN102332089 B CN 102332089B CN 201110172116 CN201110172116 CN 201110172116 CN 201110172116 A CN201110172116 A CN 201110172116A CN 102332089 B CN102332089 B CN 102332089B
- Authority
- CN
- China
- Prior art keywords
- image
- brake shoe
- fault
- shoe pricker
- pricker
- 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.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 36
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 21
- 238000012549 training Methods 0.000 claims description 10
- 210000005036 nerve Anatomy 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 8
- 238000000605 extraction Methods 0.000 claims description 7
- 239000000284 extract Substances 0.000 claims description 5
- 238000005070 sampling Methods 0.000 claims description 4
- 238000010586 diagram Methods 0.000 claims description 3
- 238000009432 framing Methods 0.000 claims description 3
- 238000007689 inspection Methods 0.000 abstract description 2
- 238000003745 diagnosis Methods 0.000 abstract 1
- 238000007781 pre-processing Methods 0.000 abstract 1
- 238000005516 engineering process Methods 0.000 description 5
- 210000002569 neuron Anatomy 0.000 description 4
- 238000001514 detection method Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000003909 pattern recognition Methods 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000004040 coloring Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 230000003137 locomotive effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
Images
Landscapes
- Image Analysis (AREA)
- Train Traffic Observation, Control, And Security (AREA)
Abstract
The invention discloses a railway wagon brake shoe key going-out fault recognition method based on an artificial neural network. The method comprises the following steps: (1) obtaining an image, model and wagon information of a running railway wagon body when a train passes through; (2) reading a to-be-recognized image; (3) performing preliminary positioning according to an image name to obtain a preliminarily positioned image; (4) scaling-down the preliminarily positioned image in an equal proportion; (5) preliminarily preprocessing the image scaled down; (6) positioning a candidate region for the preliminarily preprocessed image; (7) extracting an image feature vector of the candidate region; (8) carrying out recognition computing on the feature data; (9) outputting a recognition result. The method cannot influence the normal work of a train inspector on the existing TFDS (travelling fault diagnosis system) system; an automatic recognition result can be stored in a friendly form to provide a convenient re-inspection manner for the train inspector; the method has the advantage of full automatic start or stop and is free from manual intervention in service.
Description
Technical field
The invention belongs to railway freight-car operation troubles detection range, relate to TFDS lorry operation troubles dynamic image detection range, be specifically related to the scurry out automatic distinguishing method for image of fault of railway freight-car brake shoe pricker.
Background technology
Railway freight-car operation troubles rail edge graph is a cover collection high-speed figure image acquisition as detection system, high capacity image real-time processing technique, and placement technology, networking technology and automatic control technology are called for short TFDS in the system of one.
The TFDS system mainly carries out Fault Identification in mode artificial with the aid of pictures to the image of gathering at present, the human cost height, and labour intensity is big.The effect of Fault Identification is subjected to train-examiner's physiological status and examines the restriction of subjective factors such as car experience.The utilization of image automatic identification technology can reduce the working strength of station inspector greatly, significantly improves work efficiency, also can reduce the quantity of image station inspector, thereby reduces the operating cost of system.
The automatic Recognition Theory of image has obtained in a plurality of fields such as recognition of face, fingerprint recognition, production line safety checks using widely, but also relatively limited as the research in the automatic identification field in the lorry fault graph, this type of application also do not occur.
Neural network algorithm is a kind of intelligent algorithm of simulating human neuron activity, is widely used in fields such as System Discrimination, pattern-recognition, Based Intelligent Control.Neural network algorithm is that the basis is described with neuronic mathematical model, is represented by network topology structure, node diagnostic, learning rules.Have following several advantage:
1, parallel processing capability;
2, height robustness and fault-tolerant ability;
3, distributed storage and learning ability;
4, fully approach complicated nonlinear relationship ability.
Since occurring the forties in 20th century, neural network algorithm has experienced the development in 70 years, has formed tens of kinds of algorithm models, and BP network, Hopfield network, ART network, Kohonen network etc. are more typically arranged.
To the result for retrieval of existing document and patent, also the cutout cock handgrip is not closed the correlative study and the patent of this class fault from present.This paper method has been filled up this blank.
Summary of the invention
To the objective of the invention is in order addressing the above problem, to propose a kind of railway freight-car brake shoe pricker fault recognition method of scurrying out based on artificial neural network.
A kind of railway freight-car brake shoe pricker based on artificial neural network fault recognition method of scurrying out is characterized in that, comprises following steps:
When (1) train passes through, obtain image, vehicle and the information of vehicles of lorry body in service;
Information of vehicles is that lorry comprises vehicle number, and image and vehicle deposit the truck body image library in, and image is named, and can know whether contain vehicle braked beam position in the image by the name of image, promptly whether contains the brake shoe pricker;
(2) read in image to be identified;
From the truck body image library, extract this train according to the name of vehicle vehicle and car body image and contain the image of brake shoe pricker parts as image to be identified;
(3) carry out location just according to image name, obtain the image after the Primary Location;
Tentatively select the scope of brake shoe pricker image according to image name, obtain the image after the Primary Location from image center to be identified;
(4) picture size after the scaled down Primary Location;
Size of images after adopting image processing algorithm to Primary Location is carried out scaled down, obtains dwindling the back image;
(5) dwindle the preliminary pre-service of image afterwards;
The coloured image that at first will dwindle the back image is transformed into gray level image, then gray level image is carried out successively that brightness adjustment, contrast stretch, image is negated, and finishes the preliminary pre-service of image;
(6) to framing candidate region after the preliminary pre-service;
Calculate the horizontal direction and the vertical gradient value of preliminary pretreated image, accurately locate the position of brake shoe pricker by the size of judging both direction Grad summation, the position of brake shoe pricker is the candidate region;
(7) the characteristics of image vector of extraction candidate region;
Utilize feature extracting method that feature extraction is carried out in the candidate region, obtain characteristic;
(8) characteristic is discerned computing;
Adopt the conduct of BP neural network based on artificial nerve network classifier, obtain training sample, training sample is divided into fault sample and non-fault sample, fault sample be comprise characteristic with and corresponding fault, location of fault information, characteristic of correspondence data when non-fault sample comprises non-fault; The characteristic input of extracting from the candidate region is discerned based on artificial nerve network classifier, obtained the brake shoe pricker and whether have fault and location of fault information; The fault that described fault is scurried out for the brake shoe pricker, the position that location of fault information is scurried out for the brake shoe pricker;
(9) output recognition result;
The location of fault that the brake shoe pricker that identifies is scurried out is marked on the image, and output shows.
The invention has the advantages that:
(1) do not influence the operate as normal of station inspector in existing TFDS system;
(2) preserve automatic recognition result in close friend's mode, provide the mode of rechecking easily to station inspector;
(3) automatically start, stop, in servicely do not need manual intervention;
(4) type of vehicle is carried out careful classification, adaptive faculty is strong, also is easy to increase new type of vehicle;
(5) name of image has the predefine rule, makes automatic identification algorithm simpler and more direct.
Description of drawings
Fig. 1 is a method flow diagram of the present invention.
Among the figure:
Embodiment
The present invention is described in further detail below in conjunction with drawings and Examples.
The present invention is a kind of railway freight-car brake shoe pricker based on artificial neural network fault recognition method of scurrying out, and flow process comprises following steps as shown in Figure 1:
(1) train by the time, receive image, vehicle and the information of vehicles of the lorry body in service that sends from the high-speed image sampling device;
Information of vehicles is that lorry comprises vehicle number, and a row lorry probably is made of 40-70 car and 1 locomotive, and image and vehicle deposit the truck body image library in, and the naming rule of image is: x_y_z, x are the numbering of vehicle in the train number train, since 1 layout; Y is the position of this vehicle, and x_1_z represents the bogie position of vehicle, and x_2_z represents vehicle braked beam position, and x_3_z represents vehicle pars intermedia position, and x_4_z represents that hitch colludes slow portion; Z is the numbering of this station diagram picture of train, since 1 layout, and the vehicle model difference, its vehicle length difference, so amount of images is also different.The 8th image representing the 3rd car brake beam position of this train as 3_2_8.jpg.
(2) read in image to be identified;
From the truck body image library, extract this train according to the name of vehicle vehicle and car body image and contain the image of brake shoe pricker parts as image to be identified.
The brake shoe pricker is the parts at brake beam position, therefore the brake shoe pricker only can appear in the x_2_z image, further, this method is by big quantitative statistics, and these parts generally are present in the image of x_2_0, x_2_2, x_2_3, x_2_5, x_2_6, x_2_8, x_2_9, x_2_11;
(3) carry out location just according to image name, obtain the image after the Primary Location;
Tentatively select the approximate range of brake shoe pricker image according to image name, obtain the image after the Primary Location from image center to be identified.
Be specially: in the image of x_2_0 and x_2_6, roughly (221-784 is in scope 377-700) in original image for the brake shoe pricker; In the image of x_2_2 and x_2_8, roughly (784-1024 is in scope 700-1170) in original image for the brake shoe pricker; In the image of x_2_3 and x_2_9, roughly (194-324 is in scope 240-700) in former figure for the brake shoe pricker; In the image of x_2_5 and x_2_11, roughly (700-1024 is in scope 240-1180) in former figure for the brake shoe pricker.Wherein, (m, n) in, the row of m presentation video is high, the col width of n presentation video, (a-c, b-d) in the scope of expression m be a to c, the scope of n is that b is to d.A large amount of by analysis lorry types and consult tens million of fault graph pictures after proof be feasible with the position of this corresponding relation failure judgement candidate region.
(4) picture size after the scaled down Primary Location;
Size of images after adopting image processing algorithm to Primary Location is carried out scaled down, obtains dwindling the back image.Described image processing algorithm is the image drop sampling algorithm, and the equal proportion convergent-divergent gets scaling and is 1/4 (it is original 1/2 that length and width respectively become, and area becomes original 1/4).Recognition speed and recognition accuracy can reach best balance under this ratio.
(5) dwindle the preliminary pre-service of image afterwards;
The coloured image that at first will dwindle the back image is transformed into gray level image, then gray level image is carried out successively that brightness adjustment, contrast stretch, image is negated, and finishes the preliminary pre-service of image.
It is the notion of a Flame Image Process that contrast stretches, and generally for the difference degree of increase prospect and background, concrete method is different according to demand and design, and mainly is convenient to the extraction of brake shoe pricker component feature vector in image among the present invention.The image available formula of negating is described: and y (i, j)=255-x (i, j), y represents output image, x represents input picture, x (i, j) denotation coordination is i, the input image pixels value of j, (i, j) denotation coordination is i to y, the output image pixel value of j.
This preprocess method is changed to gray level image to image from coloured image, can accomplish to abandon colouring information in the process of conversion, but does not lose fault characteristic information.
(6) to framing candidate region after the preliminary pre-service;
Calculate the horizontal direction and the vertical gradient value of preliminary pretreated image, accurately locate the position of brake shoe pricker by the size of judging both direction Grad summation, the position of brake shoe pricker is the candidate region.
Be specially: horizontal direction and the vertical gradient value (sobel value) of obtaining image after the preliminary pre-service, by judging the size of each point in image Grad summation of horizontal direction and vertical direction in 100～200 adjacent pixel values of upper and lower, left and right four direction, when the value of summation greater than 185 the time, become candidate point, the position that the set of all candidate points constitutes the brake shoe pricker is the candidate region.
(7) the characteristics of image vector of extraction candidate region;
Utilize the various features extracting method that feature extraction is carried out in the candidate region, obtain characteristic.
Described feature extracting method is: Edge Gradient Feature, and corner characteristics extracts, and the ridge detected characteristics is extracted or provincial characteristics is extracted.On the mathematics, ridge is defined as the extreme point on the largest face curvature direction, can detect by the eigenwert of calculating the Hessian matrix.Fixed size ridge vector is highstrung to target width.It makes scale parameter along with ridge structure in the image automatically adjusts.
(8) characteristic is discerned computing;
The characteristic input of extracting from the candidate region is discerned based on artificial nerve network classifier, obtained the brake shoe pricker and whether have fault and location of fault information.
Adopt the BP neural network based on artificial nerve network classifier among the present invention, neural network algorithm is neuron models and the network topology structure that adopts, and has learning functionality, can adjust its mode of learning according to the situation of training sample, to obtain better effect.
The BP neural network is made up of the forward-propagating of information and two processes of backpropagation of error.Each neuron of input layer is responsible for receiving the input information that comes from the outside, and passes to each neuron of middle layer; The middle layer is the internal information processing layer, is responsible for information conversion, and according to the demand of information change ability, the middle layer can be designed as single hidden layer or many hidden layers structure; Last hidden layer is delivered to output layer, and each neuronic information is finished the once forward-propagating processing procedure of study after further handling, by output layer to extraneous output information result.When reality output is not inconsistent with desired output, enter the back-propagation phase of error.Error is by output layer, by each layer of mode correction weights of error gradient decline, to the anti-pass successively of hidden layer, input layer.Information forward-propagating that goes round and begins again and error back propagation process, it is the process that each layer weights are constantly adjusted, also be the process of neural network learning training, the error that this process is performed until network output reduces to the acceptable degree, till the perhaps predefined study number of times.
Described training sample is divided into fault sample and non-fault sample, all obtains from the train view data by the mode of manually choosing figure; Fault sample be comprise characteristic with and corresponding fault, location of fault information.Characteristic of correspondence data when non-fault sample comprises non-fault.
,, after training is finished the characteristic input is discerned based on artificial nerve network classifier training according to training sample, obtained the brake shoe pricker and whether have fault and location of fault information based on artificial nerve network classifier.
The fault that described fault is scurried out for the brake shoe pricker, the position that location of fault information is scurried out for the brake shoe pricker.
(9) output recognition result;
The location of fault that the brake shoe pricker that identifies is scurried out is marked on the image, and output shows.
Data result can manually directly be checked, also can be visited automatically in programmable mode by other system.Preceding a kind of artificial direct viewing of being convenient to, a kind of being convenient in back is combined in the more complicated application system of formation in the other system to this device.
Integrated application of the present invention the basic theory of artificial neural network in Flame Image Process and the pattern-recognition, and improve and integrate, the railway freight-car brake shoe pricker fault graph of scurrying out is looked like to discern, human cost and security staff's labour intensity of lorry safe operation system have been reduced, strengthened the security control ability of row inspections, for guaranteeing that transportation safety is unimpeded, improve conevying efficiency and have very important significance.
Claims (5)
1. the railway freight-car brake shoe pricker based on artificial neural network fault recognition method of scurrying out is characterized in that, comprises following steps:
When (1) train passes through, obtain image, vehicle and the information of vehicles of lorry body in service;
Information of vehicles is that lorry comprises vehicle number, and image and vehicle deposit the truck body image library in, and image is named, and can know whether contain vehicle braked beam position in the image by the name of image, promptly whether contains the brake shoe pricker;
The naming rule of image is in described (1): x_y_z, x are the numbering of vehicle in the train number train, since 1 layout; Y is the position of this vehicle, and x_1_z represents the bogie position of vehicle, and x_2_z represents vehicle braked beam position, and x_3_z represents vehicle pars intermedia position, and x_4_z represents that hitch colludes slow portion; Z is the numbering of this station diagram picture of train, since 1 layout;
(2) read in image to be identified;
From the truck body image library, extract this train according to the name of vehicle vehicle and car body image and contain the image of brake shoe pricker parts as image to be identified;
Image to be identified is x_2_0, x_2_2, x_2_3, x_2_5, x_2_6, x_2_8, x_2_9 and x_2_11 in described (2);
(3) carry out location just according to image name, obtain the image after the Primary Location;
Tentatively select the scope of brake shoe pricker image according to image name, obtain the image after the Primary Location from image center to be identified;
In described (3), be specially: in the image of x_2_0 and x_2_6, roughly (221-784 is in scope 377-700) in original image for the brake shoe pricker; In the image of x_2_2 and x_2_8, roughly (784-1024 is in scope 700-1170) in original image for the brake shoe pricker; In the image of x_2_3 and x_2_9, roughly (194-324 is in scope 240-700) in former figure for the brake shoe pricker; In the image of x_2_5 and x_2_11, the brake shoe pricker is (700-1024 in former figure roughly, in scope 240-1180), wherein, (221-784,377-700), (784-1024,700-1170), (194-324,240-700), (700-1024 240-1180) adopts (a-c, b-d) expression, a represents the pixel of brake shoe pricker initial row in original image, c represents the pixel of brake shoe pricker termination row in original image, and b represents the pixel of brake shoe pricker initial row in original image, and d represents the pixel of brake shoe pricker end column in original image;
(4) picture size after the scaled down Primary Location;
Size of images after adopting image processing algorithm to Primary Location is carried out scaled down, obtains dwindling the back image;
(5) dwindle the preliminary pre-service of image afterwards;
The coloured image that at first will dwindle the back image is transformed into gray level image, then gray level image is carried out successively that brightness adjustment, contrast stretch, image is negated, and finishes the preliminary pre-service of image;
(6) to framing candidate region after the preliminary pre-service;
Calculate the horizontal direction and the vertical gradient value of preliminary pretreated image, accurately locate the position of brake shoe pricker by the size of judging both direction Grad summation, the position of brake shoe pricker is the candidate region;
Described (6) are specially: horizontal direction and the vertical gradient value of obtaining image after the preliminary pre-service, by judging the size of each point in image Grad summation of horizontal direction and vertical direction in 100～200 adjacent pixel values of upper and lower, left and right four direction, when the value of summation greater than 185 the time, become candidate point, the position that the set of all candidate points constitutes the brake shoe pricker is the candidate region;
(7) the characteristics of image vector of extraction candidate region;
Utilize feature extracting method that feature extraction is carried out in the candidate region, obtain characteristic;
(8) characteristic is discerned computing;
Adopt the conduct of BP neural network based on artificial nerve network classifier, obtain training sample, training sample is divided into fault sample and non-fault sample, fault sample be comprise characteristic with and corresponding fault, location of fault information, characteristic of correspondence data when non-fault sample comprises non-fault; The characteristic input of extracting from the candidate region is discerned based on artificial nerve network classifier, obtained the brake shoe pricker and whether have fault and location of fault information; The fault that described fault is scurried out for the brake shoe pricker, the position that location of fault information is scurried out for the brake shoe pricker;
(9) output recognition result;
The location of fault that the brake shoe pricker that identifies is scurried out is marked on the image, and output shows.
2. a kind of railway freight-car brake shoe pricker based on artificial neural network according to claim 1 fault recognition method of scurrying out is characterized in that, image, vehicle and the information of vehicles of lorry body obtain by the high-speed image sampling device in described (1).
3. a kind of railway freight-car brake shoe pricker based on artificial neural network according to claim 1 fault recognition method of scurrying out is characterized in that, image processing algorithm is the image drop sampling algorithm in described (4).
4. a kind of railway freight-car brake shoe pricker according to claim 1 fault recognition method of scurrying out based on artificial neural network, it is characterized in that, described (4) moderate proportions is reduced into: scaling is 1/4, and it is original 1/2 that length and width respectively become, and it is original 1/4 originally that area becomes.
5. a kind of railway freight-car brake shoe pricker according to claim 1 fault recognition method of scurrying out based on artificial neural network, it is characterized in that, feature extracting method is in described (7): Edge Gradient Feature, and corner characteristics extracts, and the ridge detected characteristics is extracted or provincial characteristics is extracted.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110172116 CN102332089B (en) | 2011-06-23 | 2011-06-23 | Railway wagon brake shoe key going-out fault recognition method based on artificial neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110172116 CN102332089B (en) | 2011-06-23 | 2011-06-23 | Railway wagon brake shoe key going-out fault recognition method based on artificial neural network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102332089A CN102332089A (en) | 2012-01-25 |
| CN102332089B true CN102332089B (en) | 2013-07-24 |
Family
ID=45483860
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201110172116 Expired - Fee Related CN102332089B (en) | 2011-06-23 | 2011-06-23 | Railway wagon brake shoe key going-out fault recognition method based on artificial neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102332089B (en) |
Families Citing this family (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103499584B (en) * | 2013-10-16 | 2016-02-17 | 北京航空航天大学 | Railway wagon hand brake chain bar loses the automatic testing method of fault |
| CN107430677B (en) * | 2015-03-20 | 2022-04-12 | 英特尔公司 | Target identification based on improving binary convolution neural network characteristics |
| CN106940884B (en) * | 2015-12-15 | 2020-11-24 | 北京康拓红外技术股份有限公司 | Motor train unit operation fault image detection system containing depth information and detection method |
| CN106600581B (en) * | 2016-12-02 | 2019-09-24 | 北京航空航天大学 | A kind of train operation automatic fault detection system and method based on binocular stereo vision |
| CN106778740A (en) * | 2016-12-06 | 2017-05-31 | 北京航空航天大学 | A kind of TFDS non-faulting image detecting methods based on deep learning |
| CN107273802B (en) * | 2017-05-16 | 2020-05-22 | 武汉华目信息技术有限责任公司 | Method and device for detecting fault of brake shoe drill rod ring of railway train |
| CN108937726A (en) * | 2017-11-27 | 2018-12-07 | 炬大科技有限公司 | A kind of clean robot awakening method and device based on cleannes identification |
| CN109949269A (en) * | 2019-01-24 | 2019-06-28 | 武汉华目信息技术有限责任公司 | A kind of detection method and device of railroad train dust cap breakage failure |
| CN111091546B (en) * | 2019-12-12 | 2020-10-30 | 哈尔滨市科佳通用机电股份有限公司 | Railway wagon coupler tail frame breaking fault identification method |
| CN111080608A (en) * | 2019-12-12 | 2020-04-28 | 哈尔滨市科佳通用机电股份有限公司 | Method for recognizing closing fault image of automatic brake valve plug handle of railway wagon in derailment |
| CN111091542B (en) * | 2019-12-12 | 2020-11-27 | 哈尔滨市科佳通用机电股份有限公司 | Image identification method for breakage fault of spring supporting plate of railway wagon bogie |
| CN111091558B (en) * | 2019-12-12 | 2020-08-21 | 哈尔滨市科佳通用机电股份有限公司 | Railway wagon swing bolster spring jumping fault image identification method |
| CN111080605A (en) * | 2019-12-12 | 2020-04-28 | 哈尔滨市科佳通用机电股份有限公司 | Method for identifying railway wagon manual brake shaft chain falling fault image |
| CN111830934A (en) * | 2020-07-17 | 2020-10-27 | 中车大连电力牵引研发中心有限公司 | Electric locomotive fault source positioning method and device |
| CN112085723B (en) * | 2020-09-09 | 2021-04-02 | 哈尔滨市科佳通用机电股份有限公司 | Automatic detection method for spring jumping fault of truck bolster |
| CN112733940B (en) * | 2021-01-12 | 2021-08-27 | 哈尔滨市科佳通用机电股份有限公司 | Cut-off cock handle closing fault image identification method |
| CN113705494A (en) * | 2021-09-01 | 2021-11-26 | 武汉铁路职业技术学院 | Dynamic detection method, system and device for braking performance of freight train |
| CN116188449B (en) * | 2023-03-13 | 2023-08-08 | 哈尔滨市科佳通用机电股份有限公司 | Rail wagon relief valve pull rod split pin loss fault identification method and equipment |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102072789A (en) * | 2010-11-03 | 2011-05-25 | 西南交通大学 | Method for continuously processing wheel-track force of track vehicle by means of ground test |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| IL117279A (en) * | 1996-02-27 | 2000-01-31 | Israel Aircraft Ind Ltd | System for detecting obstacles on a railway track |
-
2011
- 2011-06-23 CN CN 201110172116 patent/CN102332089B/en not_active Expired - Fee Related
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102072789A (en) * | 2010-11-03 | 2011-05-25 | 西南交通大学 | Method for continuously processing wheel-track force of track vehicle by means of ground test |
Non-Patent Citations (3)
| Title |
|---|
| 《BP神经网络和遗传算法在货车车锁检测中的应用》;严柏军等;《计算机辅助设计与图形学学报》;20021231;第14卷(第12期);正文第1179-1183页 * |
| US 6,163,755 A,2000.12.19,说明书全文. |
| 严柏军等.《BP神经网络和遗传算法在货车车锁检测中的应用》.《计算机辅助设计与图形学学报》.2002,第14卷(第12期),正文第1179-1183页. |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102332089A (en) | 2012-01-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102332089B (en) | Railway wagon brake shoe key going-out fault recognition method based on artificial neural network | |
| CN106611169B (en) | A kind of dangerous driving behavior real-time detection method based on deep learning | |
| CN110399816B (en) | High-speed train bottom foreign matter detection method based on Faster R-CNN | |
| CN101713776B (en) | Neural network-based method for identifying and classifying visible components in urine | |
| CN105260744B (en) | The automatic on-line diagnostic method and system of a kind of goods train coupler yoke key position failure | |
| CN103258213B (en) | A kind of for the dynamic vehicle model recognizing method in intelligent transportation system | |
| CN107085696A (en) | A kind of vehicle location and type identifier method based on bayonet socket image | |
| CN110163187A (en) | Remote road traffic sign detection recognition methods based on F-RCNN | |
| CN107220603A (en) | Vehicle checking method and device based on deep learning | |
| CN104077577A (en) | Trademark detection method based on convolutional neural network | |
| CN110321891A (en) | A kind of big infusion medical fluid foreign matter object detection method of combined depth neural network and clustering algorithm | |
| CN102346844B (en) | Device and method for identifying fault of losing screw bolts for truck center plates | |
| CN104268588B (en) | Railway wagon brake shoe pricker loses the automatic testing method of failure | |
| CN104517122A (en) | Image target recognition method based on optimized convolution architecture | |
| CN110120036A (en) | A kind of multiple dimensioned tire X-ray defect detection method | |
| CN103279759A (en) | Vehicle front trafficability analyzing method based on convolution nerve network | |
| CN100418469C (en) | Computerized disease sign analysis system based on tongue picture characteristics | |
| CN103295027A (en) | Freight wagon blocking key missing fault identification method based on support vector machine | |
| CN114399672A (en) | Railway wagon brake shoe fault detection method based on deep learning | |
| CN111862029A (en) | Fault detection method for bolt part of vertical shock absorber of railway motor train unit | |
| CN106934380A (en) | A kind of indoor pedestrian detection and tracking based on HOG and MeanShift algorithms | |
| CN109935080A (en) | The monitoring system and method that a kind of vehicle flowrate on traffic route calculates in real time | |
| CN108509950A (en) | Railway contact line pillar number plate based on probability characteristics Weighted Fusion detects method of identification | |
| CN105809219B (en) | A kind of the prefabricated pipe section quality testing statistical system and method for tunnel prefabricated pipe section production line | |
| CN109522831A (en) | A kind of vehicle real-time detection method based on micro- convolutional neural networks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20130724 |