CN102323070A - Method and system for detecting abnormality of train - Google Patents
Method and system for detecting abnormality of train Download PDFInfo
- Publication number
- CN102323070A CN102323070A CN201110155792A CN201110155792A CN102323070A CN 102323070 A CN102323070 A CN 102323070A CN 201110155792 A CN201110155792 A CN 201110155792A CN 201110155792 A CN201110155792 A CN 201110155792A CN 102323070 A CN102323070 A CN 102323070A
- Authority
- CN
- China
- Prior art keywords
- image
- train
- reference picture
- current train
- current
- 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
Images
Landscapes
- Image Analysis (AREA)
Abstract
The invention provides a method and system for detecting abnormality of a train. The method comprises the following steps of: acquiring a global image of the current train; determining a number of the current train and selecting a global image corresponding to the number of the current train from a preset image library as a reference image; and aligning and comparing the global image of the current train with the reference image so as to determine a region of the global image of the current train, which is inconsistent with the reference image, as an abnormal region of the train. According to the scheme provided by the embodiment of the invention, the detection of an actual vehicle is converted into the analysis of the image, thereby automatic comparison can be carried out by adopting a computer-assisted mode and the problems of low efficiency and high possibility of leakage detection caused by mainly depending manual labor in the prior art are solved.
Description
Technical field
The present invention relates to technical field of traffic transportation, relate to a kind of train method for detecting abnormality and system in particular.
Background technology
Transportation by railroad is big with freight volume, quick, security reliability advantages of higher, in technical field of traffic transportation in occupation of the position of outbalance.
And train is as the core of transportation by railroad, its abnormality detection comprehensively, accurately, guarantee is vital for safety of railway traffic fast.
But present train abnormality detection mode mainly is that the staff rule of thumb investigates, and there is following shortcoming at least in this mode:
1, train (comprising lorry, passenger vehicle, EMUs and other type column cars) is formed complex structure, and tiny parts are more, adopts the manual detection mode to have the problem of inefficiency and easy omission;
2, many factors such as the method for operation all can increase the difficulty of manual detection, have further reduced work efficiency and have further increased the probability of omission:
With the motor-car is example, and platform height is higher now, causes the hidden number of components of train to increase; It is bigger with traditional train difference that it forms structure, and upkeep operation personnel difficulty remember the normal condition form of each parts; In addition, behind warehouse-in, the interior outer rim of its wheel, tread, wheel rim are because blocking of rail and bogie structure exists the vision blind area.In addition, the EMUs operation characteristic is: one stands erectly reaches, the dwell time short and the long routing operation, and these characteristics cause it to overhaul with manual type midway.
Therefore need utilize the automatic abnormality detection mode of computing machine indirect labor to detect, the reduction work difficulty is also increased work efficiency.In addition; Because the train component number is a lot of, fault type is difficult to calculate, and the existing fault detection method is difficult to set up accurate and complete fault model; So; Described train fault detection method is through the global image comparison zone that notes abnormalities, and the mode of the further recognition and verification of binding key fault is to abnormal area classification, classification, and the legend alarm.
Summary of the invention
In view of this, the present invention provides a kind of train method for detecting abnormality and system, and the inefficiency of bringing in order to solution prior art manual detection mode reaches the problem that occurs omission easily.
For realizing above-mentioned purpose, the present invention provides following technical scheme:
A kind of train method for detecting abnormality comprises:
Obtain current train global image;
Confirm current train license number, in preset image library, select and the corresponding global image of current train license number image as a reference;
With the comparison of aliging with said reference picture of said current train global image, be the train abnormal area with the inconsistent zone of confirming said current train global image and said reference picture.
Preferably, said method also comprises:
Said train abnormal area is carried out the legend sign.
Preferably, in the said method, saidly said train abnormal area carried out the legend sign be specially:
Said train abnormal area is carried out classification according to the order of severity, and wherein the severely subnormal zone utilizes the conventional fault detection method to carry out the classification of emphasis Fault Identification, and different brackets, different classes of abnormal area indicate with different colours or shape and show.
Preferably, said method also comprises:
According to preset grade setting standard, confirm the rank of abnormal area;
Unusual rank is carried out the legend demonstration greater than the abnormal area of presetting thresholding.
Preferably, said method also comprises:
Receiving after the user cancels the indication of abnormal area the abnormal area of the said user's appointment of cancellation in said current global image.
Preferably, said method also comprises, deposits said current train global image in said preset image library.
Preferably, in the said method, saidly obtain current train image and be specially: utilize linear array or face battle array imaging mode to obtain current train global image.
Preferably, in the said method, the corresponding reference picture of said and current train license number is:
The image of the normal train that is provided with in advance, perhaps,
With the image of the most contiguous same the car that passes through of current time, perhaps,
With many global images of contiguous same the car that passes through of current time, perhaps,
By said many global images with contiguous same the car that passes through of current time merge or statistical study after the reference picture that obtains.
Preferably, in the said method, said said current train global image is alignd with said reference picture comprises:
For face system of battle formations picture, adopt the mode of global registration;
For linear array images, detect current train speed, and inquiry obtains the corresponding train speed of said reference picture;
When the difference of the said current train speed train speed corresponding during less than default value with said reference picture; Utilize the mode of unique point global registration that said current train image and said reference picture are alignd; Otherwise, said current train image and said reference picture are alignd with local registration mode.
Preferably, in the said method, the mode of said unique point global registration comprises:
Utilize the SIFT/SURF method to try to achieve a plurality of unique points of said current train image and reference picture; And the proper vector that constitutes of yardstick and the direction of preserving each unique point; Utilize the Euclidean distance method to find out unique point characteristic of correspondence point in said reference picture of current train image respectively, it is right to constitute same place;
It is right to reject erroneous point according to the RANSAC algorithm according to projective rejection;
Confirm the coordinate transform mapping relations of the same place of RANSAC reservation to correspondence;
Reference picture is carried out interpolation arithmetic.
Preferably, in the said method, the mode of said local registration comprises:
Utilize the SIFT/SURF method to try to achieve a plurality of unique points of said current train image and reference picture; And the proper vector that constitutes of yardstick and the direction of preserving each unique point; Utilize the Euclidean distance method to find out unique point characteristic of correspondence point in said reference picture of current train image respectively, it is right to constitute same place;
Utilize projective transformation to confirm that each same place carries out interpolation arithmetic to the coordinate transform mapping relations of correspondence to reference picture successively.
Preferably, in the said method, said comparison comprises:
Confirm the marginal portion of current train image and reference picture, utilize marginal information to compare.
Preferably, in the said method, before the marginal portion of confirming current train image and reference picture, also comprise:
Utilize the statistics with histogram method that current train image and reference picture are carried out brightness normalization processing.
Preferably, in the said method, after the marginal portion of confirming current train image and reference picture, also comprise: the edge of current train image and reference picture is carried out the normalization processing.
The present invention also discloses a kind of train abnormality detection system simultaneously, comprising:
Image acquisition unit is used to obtain current train global image;
The train license number is confirmed the unit, is used for confirming current train license number;
Reference picture is chosen the unit, is used for selecting and the corresponding global image of current train license number image as a reference in preset image library;
The alignment comparing unit is used for the comparison of aliging with said reference picture of said current train global image;
The train abnormal area is confirmed the unit; Be used for alignment comparison result according to said alignment comparing unit; The inconsistent zone of confirming said current train global image and said reference picture is the train abnormal area; And, severely subnormal is carried out the classification of emphasis Fault Identification according to the order of severity of abnormal area.
Preferably, said system also comprises:
The legend unit is used for said train abnormal area is carried out the legend sign, and perhaps, when having a plurality of abnormal area, the abnormal area that rank is surpassed preset thresholding carries out the legend sign.
Preferably, said system also comprises:
Image deposits the unit in, is used for depositing said current train global image in said preset image library.
Can know via above-mentioned technical scheme; Compared with prior art; The scheme that present embodiment provides will be transformed into the analysis to image to the detection of actual vehicle; Thereby can adopt computer-assisted way to compare automatically, solve prior art and too much rely on manual work and inefficiency that causes and the problem that occurs omission easily.
Description of drawings
In order to be illustrated more clearly in the embodiment of the invention or technical scheme of the prior art; To do to introduce simply to the accompanying drawing of required use in embodiment or the description of the Prior Art below; Obviously, the accompanying drawing in describing below only is embodiments of the invention, for those of ordinary skills; Under the prerequisite of not paying creative work, can also obtain other accompanying drawing according to the accompanying drawing that provides.
A kind of train method for detecting abnormality process flow diagram that Fig. 1 provides for the embodiment of the invention;
The particular flow sheet of step S14 in a kind of train method for detecting abnormality process flow diagram that Fig. 2 provides for the embodiment of the invention;
The particular flow sheet of step S15 in a kind of train method for detecting abnormality process flow diagram that Fig. 3 provides for the embodiment of the invention;
The another kind of train method for detecting abnormality process flow diagram that Fig. 4 provides for the embodiment of the invention;
Another train method for detecting abnormality process flow diagram that Fig. 5 provides for the embodiment of the invention;
The structural representation of a kind of train abnormality detection system that Fig. 6 provides for the embodiment of the invention;
The structural representation of the another kind of train abnormality detection system that Fig. 7 provides for the embodiment of the invention;
The structural representation of another train abnormality detection system that Fig. 8 provides for the embodiment of the invention;
The structural representation of another train abnormality detection system that Fig. 9 provides for the embodiment of the invention;
The structural representation of another train abnormality detection system that Figure 10 provides for the embodiment of the invention.
Embodiment
To combine the accompanying drawing in the embodiment of the invention below, the technical scheme in the embodiment of the invention is carried out clear, intactly description, obviously, described embodiment only is the present invention's part embodiment, rather than whole embodiment.Based on the embodiment among the present invention, those of ordinary skills are not making the every other embodiment that is obtained under the creative work prerequisite, all belong to the scope of the present invention's protection.
The embodiment of the invention discloses a kind of train method for detecting abnormality; It mainly is to confirm unusual through the mode of image comparison; To be transformed into analysis to the detection of actual vehicle to image; Thereby can adopt computer-assisted way to compare automatically, solve prior art and too much rely on manual work and inefficiency that causes and the problem that occurs omission easily.The substance of this method is: through obtaining current train global image automatically; And definite current train license number; Then; In preset image library, select and the corresponding global image of current train license number image as a reference, said current global image and reference picture are compared, whether finally definite current train occurs unusually.
Through several embodiment technical scheme of the present invention is elaborated below.
Embodiment one
A kind of train method for detecting abnormality process flow diagram that present embodiment provides comprises the steps:
Step S11, obtain current train global image.
Said global image refers to the train image that photographs from all angles (like bottom, top, left side, right side etc.).
The concrete mode of obtaining current train global image can be: through being arranged on a plurality of camera picked-ups of certain predeterminated position (the next sensing point that is referred to as) in the train driving path in advance.
Its picked-up mode can adopt the imaging mode of linear array or face battle array.
Step S12, confirm current train license number.
Said current train license number specifically be meant train each the joint () compartment, every joint compartment all has unique numbering, like ZE210103.
Step S13, in preset image library, select and the corresponding global image of current type of train image as a reference.
Said reference picture can be with in the hypograph any one:
1, fixed form image
The fixed form image, the standard picture of taking when promptly train dispatches from the factory, related vehicle number information during storage.Can current vehicle image (promptly when the front compartment image) be compared with the fixed form image.
2, a normal picture of the time arest neighbors when taking said current train global image), said normal picture is meant through alignment comparison back confirms that train is no abnormal, and the concrete mode of said alignment comparison will be introduced hereinafter in detail.In general, the time is contiguous more, and reference value is maximum, and the credible result degree of itself and present image compare of analysis is the highest.
The image that photographs when 3, current train crosses certain sensing point in the most contiguous time;
The image that photographs when 4, current train crosses other sensing points in the most contiguous time.
5, said current train is being close to many images that the current time is photographed.
Select many time neighbours' image and present image difference compare of analysis, comparison result according to different weight fusion treatment, is reduced false drop rate and loss.
6, many images that are photographed in the contiguous current time based on said current train pass through the image that obtains behind the statistical mathematics models.
Step S14, said current train global image is alignd with said reference picture.
Image " alignment ", process is similar to image registration, is the prerequisite that solves change-detection.Image alignment is exactly the public scene part of finding out in two width of cloth images; And confirm the transformation parameter between them; This two width of cloth image is normally taken in different time, different illumination conditions, based on different resolution or different angles and position etc.; Conversion between them can be a rigid body translation, also can nonlinear affined transformation etc.
Detect application to train fault; For face system of battle formations picture, adopt the mode of overall situation alignment, for linear array images; " alignment " mainly solves the linear deformation problem; And mainly defer to two kinds of patterns and carry out, the one, when being more or less the same with current train speed, directly linear array images is utilized that unique point is overall aligns with reference to train; The 2nd, when when big,, introducing respectively below then to the linear array images local alignment with reference to train speed and current vehicle speed gap:
Pattern one: overall situation alignment
The main process flow diagram as shown in Figure 2 of " alignment " (above-mentioned steps S14) step in the overall situation alignment pattern is carried out, and flow process shown in Figure 2 may further comprise the steps:
Step S21, feature point extraction.
In the present embodiment, mainly adopt SIFT/SURF and combine harris isocenter detection algorithm to carry out weighting scheme keeping characteristics point by different weights.
Concrete mode is: at first; Utilize the SIFT algorithm to carry out feature detection at metric space, and the position of definite key point and the residing yardstick of key point, then; Use the direction character of the principal direction of key point neighborhood gradient, to realize the independence of operator to yardstick and direction as this point.
The SIFT algorithm can be handled the characteristic matching problem that takes place between two width of cloth images under translation, rotation, dimensional variation, the illumination variation situation, and can also possess comparatively stable characteristics matching capacity to visual angle change, affined transformation to a certain extent.
The SURF algorithm is the acceleration version of SIFT algorithm, and through integral image haar differentiate, the SURF algorithm is accomplished object in two width of cloth images under moderate condition coupling has realized real-time processing basically.
The Harris Corner Detection Algorithm grows up on Moravec algorithm basis.The thought of Moravec Corner Detection Algorithm is: local detection window of design in image; When this window when all directions are made minute movement; The average energy of investigating window changes, and when this energy change value surpassed preset threshold, just the central pixel point with window was extracted as angle point.It is detection window that the Harris detection algorithm is chosen Gaussian function, and image is carried out extracting angle point again behind the smothing filtering, and noise is had certain inhibiting effect.
Step S22, utilize the RANSAC method to reject erroneous point.
Can obtain yardstick, the direction constitutive characteristic vector of each unique point when trying to achieve unique point among the step S21, utilize the Euclidean distance method to find out unique point characteristic of correspondence point in reference picture of present image respectively, it is right to constitute same place.
The RANSAC algorithm calculates the mathematical model parameter of data according to one group of sample data collection that comprises abnormal data, obtains the effective sample data.In the present embodiment, at first detect unique point, by RANSAC the mistake coupling is rejected again by SIFT/SURF.
In utilizing unique point alignment, the projective rejection of the unique point on model is from the unique point on the plane to the another one plane is reacted and is projection matrix H.H is 3 * 3 matrixes that comprise 8 degree of freedom, its minimum can being calculated by 4 pairs of match points in two planes, but the conllinear not of 3 points on the same plane.
Step S23, calculating coordinate change mapping function.
It is right that RANSAC is rejected the unique point that retains after the erroneous point, and it is right to require to keep four same places at least this moment, utilizes projective transformation to try to achieve coordinate transform relation between present image and the reference picture.
Step S24, interpolation arithmetic.
Reference picture is carried out coordinate transform and interpolation processing according to the projective transform matrix that step S33 tries to achieve.Here consider interpolation and operation efficiency, adopt bilinear interpolation method.On mathematics, bilinear interpolation is the linear interpolation expansion that the interpolating function of two variablees is arranged, and its core concept is to carry out the once linear interpolation respectively at both direction.
Pattern two: local alignment
When with reference to train speed and current vehicle speed gap when big; Explanation might be when twice be gathered the train image, and speed change appears in train, and conventional overall projective transformation is difficult to try to achieve coordinate Mapping relation accurately; In this case; Can adopt the local alignment pattern, the particular content of this pattern comprises: utilize the SIFT/SURF method to try to achieve a plurality of unique points of said current train image and reference picture, and preserve the yardstick of each unique point and the proper vector that direction constitutes; Utilize the Euclidean distance method to find out unique point characteristic of correspondence point in said reference picture of current train image respectively, it is right to constitute same place; Utilize projective transformation to confirm that each same place carries out interpolation arithmetic to the coordinate transform mapping relations of correspondence to history image successively.
Can find out, compare that whole thinking is consistent, but has lacked step S22 with overall alignment pattern.
Can confirm to adopt the alignment thereof of which kind of pattern according to comparative result in more current train speed and after with reference to train speed:
When the difference of the said current train speed train speed corresponding during less than default value with said reference picture; Utilize the mode (being above-mentioned pattern one) of unique point global registration that said current train image and said reference picture are alignd; Otherwise, said current train image and said reference picture are alignd with the mode (being above-mentioned pattern two) of local registration.
Said current global image and reference picture after step S15, the comparison alignment.
Behind current vehicle image and the corresponding reference image alignment, carry out image " comparison ",, find the train fault generation area through the method for detected image differences.
Step S16, confirm that the inconsistent zone of said current train global image and said reference picture is the train abnormal area.
Can find out; The scheme that present embodiment provides will be transformed into the analysis to image to the detection of actual vehicle; Thereby can adopt computer-assisted way to compare automatically, solve prior art and too much rely on manual work and inefficiency that causes and the problem that occurs omission easily.
In the said method, the described image comparison process of step S15 mainly by following flow performing, is seen Fig. 3, comprises the steps:
Step S31, edge extracting.
Extract the fringe region of present image and reference picture respectively.
Step S32, the fringe region of said present image and reference picture is compared.
The present embodiment comparison that edge of image is regional is equal to the integral body comparison of image, practices thrift comparison time, raises the efficiency.
Need to prove,, directly carry out the image comparison, can influence comparison result, cause wrong report, therefore in other embodiments, cut down illumination effect through following dual mode owing to be easy to receive illumination effect in present image and the reference picture imaging process:
Pre-service pattern: utilize the statistics with histogram method that present image and reference picture are carried out brightness normalization processing, then compare;
Real-time tupe: the real-time normalization of edge/gray scale/texture that present image and reference picture are tried to achieve is handled, carried out the difference comparison then, reach the purpose of illumination effect removal.
In addition; Consider that edge difference only represents main difference, though representative lack comprehensive, therefore; Can detect outside the edge difference; Further detecting the gray scale/textural characteristics difference of present image and reference picture, is main to detect edge difference promptly also, and it is that the thinking of assisting is confirmed the image change zone that the gray scale of present image and reference picture/textural characteristics difference detects.
In other embodiment, can also said abnormal area be identified, specifically as shown in Figure 4 so that the operator can understand the position and the quantity of abnormal area easily, comprise following flow process:
Step S11 among step S41～step S46 and above-mentioned Fig. 1～step S16 content is basic identical;
Step S47, said abnormal area is identified back output.
Said abnormal area identified specifically can adopt variety of way (for example, with color or add form such as frame) to carry out, get final product so long as can distinguish abnormal area to come with all the other zones.
In addition; In other embodiments; Can also carry out classification to the influence degree of current train safe to abnormal area according to the abnormal area position; And give different signs to the abnormal area of different stage, for example identify the highest grade abnormal area, with yellow sign grade time high abnormal area with redness.So, operating personnel can confirm the state of current train quickly and intuitively according to the difference of sign.
Need to prove; Abnormal area occurring is not just to represent that necessarily current train occurs unusually; In the present embodiment; What the result who compares to image judged unusual and the main foundation of reporting to the police can be the amplitude (i.e. the amplitude of variation of two width of cloth image corresponding regions) of region of variation and the area size of region of variation, thinks then that greater than a certain setting value this zone is unusually when satisfying region of variation amplitude or area.
Can find out that present embodiment shows the mode of abnormal area with legendization, can give operating personnel with indication directly perceived, vivid, clear and definite, can improve operating personnel and discern unusual efficient.
Certainly; Adopt different identification to represent that the abnormal area of different brackets can let operating personnel understand current train situation intuitively and easily; When still needing operating personnel that the relation of various grades and sign is remembered; When grade classification is careful, need operating personnel to remember the relation of a lot of grades and sign, this makes troubles to the judgement of current train status to operating personnel.For this reason, the present invention other embodiment provide a kind of scheme, automatically to judging unusually and when appearance is unusual, report to the police that detailed process is as shown in Figure 5, comprises following flow process:
Step S11 among step S51～step S56 and above-mentioned Fig. 1～step S16 content is basic identical.
Step S57, judge that whether the order of severity of abnormal area surpasses preset thresholding, if, get into step S58, otherwise, process ends;
Concrete; Whether the order of severity of judging abnormal area surpasses preset thresholding and specifically can be: utilize the conventional fault detection method to judge the current abnormal area unit failure of whether attaching most importance to; Perhaps; Whether the amplitude (i.e. the amplitude of variation of two width of cloth image corresponding regions) of judging region of variation surpasses predetermined threshold, judges perhaps whether the area of region of variation surpasses predetermined threshold etc.
Step S58, report to the police.
The form of reporting to the police can have a lot, for example reports to the police through sound, image or motion (vibration), and specifically warning form belongs to prior art, repeats no more at this.
If abnormal area do not occur or lower grade of abnormal area, can think that then current train is in normal condition, in this case, accessed current train global image can be used as the reference picture of follow-up same column train abnormality detection.That is: when said abnormal area quantity be zero or the unusual rank of said abnormal area when being lower than preset thresholding, deposit said current train global image in said preset image library.
In addition, in a further embodiment, operating personnel can handle according to the comparison result that present embodiment provides: whether the artificial judgment abnormal area really breaks down; If; Then place under repair, and after repairing finishes, send the cancellation instruction through man-machine interaction unit and cancel corresponding abnormal area, otherwise; Then directly send the corresponding abnormal area of cancellation instruction cancellation, and then deposit said current global image in preset image library through man-machine interaction unit.
Certainly, can also directly deposit current global image in preset image library, image as a reference.In this case, as want the subsequent operation personnel to access said global image again, the train of correspondence is investigated unusually, and upgrade said global image, that is: will get rid of unusual abnormal area cancellation according to the investigation result.
So, when calling said global image next time, can be on the one hand with the normal region with reference to comparing, can monitor whether being repaired unusually above it on the other hand, perhaps whether become more serious, solving so that take measures.
For aforesaid each method embodiment; For simple description; So it all is expressed as a series of combination of actions, but those skilled in the art should know that the present invention does not receive the restriction of described sequence of movement; Because according to the present invention, some step can adopt other orders or carry out simultaneously.Secondly, those skilled in the art also should know, the embodiment described in the instructions all belongs to preferred embodiment, and related action and module might not be that the present invention is necessary.
Train method for detecting abnormality to the preamble proposition; The present invention also provides a kind of train abnormality detection system of realizing this method; Its a kind of concrete structure is as shown in Figure 6; Comprise: image acquisition unit 61, train license number confirm that unit 62, reference picture are chosen unit 63, alignment comparing unit 64 and train abnormal area are confirmed unit 65, wherein:
The train license number is confirmed unit 62, is used for confirming current type of train;
Reference picture is chosen unit 63, is used for selecting and the corresponding global image of current type of train image as a reference in preset image library;
The train abnormal area is confirmed unit 65; Be used for alignment comparison result according to said alignment comparing unit; The inconsistent zone of confirming said current train global image and said reference picture is the train abnormal area; And, severely subnormal is carried out the classification of emphasis Fault Identification according to the order of severity of abnormal area.
Concrete mode, the train license number that image acquisition unit 61 obtains current train global image confirms that unit 62 confirms that concrete mode, the reference picture of current type of train choose unit 63 and choose the process that the mode of reference picture, alignment comparing unit 64 are carried out alignment, comparison step; And the train abnormal area confirms that the concrete mode of unit 65 definite train abnormal areas please refer to the content of preamble method part, repeats no more at this.
The another kind of concrete structure of train abnormality detection system is as shown in Figure 7; Comprise that image acquisition unit 71, train license number confirm that unit 72, reference picture are chosen unit 73, alignment comparing unit 74, train abnormal area are confirmed unit 75 and legend unit 76, wherein:
The another kind of concrete structure of train abnormality detection system is as shown in Figure 8; Comprise that image acquisition unit 81, train license number confirm that unit 82, reference picture are chosen unit 83, alignment comparing unit 84, train abnormal area are confirmed unit 85, legend unit 86 and alarm unit 87, wherein:
The another kind of concrete structure of train abnormality detection system is as shown in Figure 9; Comprise that image acquisition unit 91, train license number confirm that unit 92, reference picture are chosen unit 93, alignment comparing unit 94, train abnormal area confirm that unit 95, legend unit 96, alarm unit 97 and image deposit unit 98 in, wherein:
In addition; Present embodiment can also comprise man-machine interaction unit; Shown in figure 10; Present embodiment comprises that image acquisition unit 101, train license number confirm that unit 102, reference picture are chosen unit 103, alignment comparing unit 104, train abnormal area confirm that unit 105, legend unit 106, alarm unit 107, image deposit unit 108 and man-machine interaction unit 109 in, wherein:
Said man-machine interaction unit 109 will pass through said train abnormal area on the one hand and confirm that unit 85 definite current global images are shown to operating personnel; Can also receive said operating personnel on the other hand train is investigated the indication information that send the back unusually; Abnormal area in the said global image upgrades, and will get rid of unusual abnormal area cancellation that is:.
The train abnormality detection system that present embodiment provides carries out the comparison alignment of the obtaining of image, image and all can carry out automatically, need not too much manual work and participates in, and has improved work efficiency and has reduced the probability of omission.
For the convenience of describing, be divided into various unit with function when describing above the device and describe respectively.Certainly, when implementing the application, can in same or a plurality of softwares and/or hardware, realize the function of each unit.
Need to prove that each embodiment adopts the mode of going forward one by one to describe in this instructions, what each embodiment stressed all is and the difference of other embodiment that identical similar part is mutually referring to getting final product between each embodiment.For the disclosed device of embodiment, because it is corresponding with the embodiment disclosed method, so description is fairly simple, relevant part is partly explained referring to method and is got final product.
In addition; Need to prove; In this article; Relational terms such as first and second grades only is used for an entity or operation are made a distinction with another entity or operation, and not necessarily requires or hint relation or the order that has any this reality between these entities or the operation.And; Term " comprises ", " comprising " or its any other variant are intended to contain comprising of nonexcludability; Thereby make and comprise that process, method, article or the equipment of a series of key elements not only comprise those key elements; But also comprise other key elements of clearly not listing, or also be included as this process, method, article or equipment intrinsic key element.Under the situation that do not having much more more restrictions, the key element that limits by statement " comprising ... ", and be not precluded within process, method, article or the equipment that comprises said key element and also have other identical element.
To the above-mentioned explanation of the disclosed embodiments, make this area professional and technical personnel can realize or use the present invention.Multiple modification to these embodiment will be conspicuous concerning those skilled in the art, and defined General Principle can realize under the situation that does not break away from the spirit or scope of the present invention in other embodiments among this paper.Therefore, the present invention will can not be restricted to these embodiment shown in this paper, but will meet and principle disclosed herein and features of novelty the wideest corresponding to scope.
Claims (17)
1. a train method for detecting abnormality is characterized in that, comprising:
Obtain current train global image;
Confirm current train license number, in preset image library, select and the corresponding global image of current train license number image as a reference;
With the comparison of aliging with said reference picture of said current train global image, be the train abnormal area with the inconsistent zone of confirming said current train global image and said reference picture.
2. the method for claim 1 is characterized in that, also comprises:
Said train abnormal area is carried out the legend sign.
3. method as claimed in claim 2 is characterized in that, saidly said train abnormal area is carried out the legend sign is specially:
Said train abnormal area is carried out classification according to the order of severity, and wherein the severely subnormal zone utilizes the conventional fault detection method to carry out the classification of emphasis Fault Identification, and different brackets, different classes of abnormal area indicate with different colours or shape and show.
4. the method for claim 1 is characterized in that, also comprises:
According to preset grade setting standard, confirm the rank of abnormal area;
Unusual rank is carried out the legend demonstration greater than the abnormal area of presetting thresholding.
5. the method for claim 1 is characterized in that, also comprises:
Receiving after the user cancels the indication of abnormal area the abnormal area of the said user's appointment of cancellation in said current global image.
6. method as claimed in claim 5 is characterized in that, also comprises, deposits said current train global image in said preset image library.
7. like any described method of claim 1-6, it is characterized in that, saidly obtain current train image and be specially: utilize linear array or face battle array imaging mode to obtain current train global image.
8. method as claimed in claim 7 is characterized in that, the corresponding reference picture of said and current train license number is:
The image of the normal train that is provided with in advance, perhaps,
With the image of the most contiguous same the car that passes through of current time, perhaps,
With many global images of contiguous same the car that passes through of current time, perhaps,
By said many global images with contiguous same the car that passes through of current time merge or statistical study after the reference picture that obtains.
9. the method for claim 1 is characterized in that, said said current train global image is alignd with said reference picture comprises:
For face system of battle formations picture, adopt the mode of global registration;
For linear array images, detect current train speed, and inquiry obtains the corresponding train speed of said reference picture;
When the difference of the said current train speed train speed corresponding during less than default value with said reference picture; Utilize the mode of unique point global registration that said current train image and said reference picture are alignd; Otherwise, said current train image and said reference picture are alignd with local registration mode.
10. method as claimed in claim 9 is characterized in that, the mode of said unique point global registration comprises:
Utilize the SIFT/SURF method to try to achieve a plurality of unique points of said current train image and reference picture; And the proper vector that constitutes of yardstick and the direction of preserving each unique point; Utilize the Euclidean distance method to find out unique point characteristic of correspondence point in said reference picture of current train image respectively, it is right to constitute same place;
It is right to reject erroneous point according to the RANSAC algorithm according to projective rejection;
Confirm the coordinate transform mapping relations of the same place of RANSAC reservation to correspondence;
Reference picture is carried out interpolation arithmetic.
11. method as claimed in claim 9 is characterized in that, the mode of said local registration comprises:
Utilize the SIFT/SURF method to try to achieve a plurality of unique points of said current train image and reference picture; And the proper vector that constitutes of yardstick and the direction of preserving each unique point; Utilize the Euclidean distance method to find out unique point characteristic of correspondence point in said reference picture of current train image respectively, it is right to constitute same place;
Utilize projective transformation to confirm that each same place carries out interpolation arithmetic to the coordinate transform mapping relations of correspondence to reference picture successively.
12. method as claimed in claim 9, said comparison comprises:
Confirm the marginal portion of current train image and reference picture, utilize marginal information to compare.
13. method as claimed in claim 12 is characterized in that, before the marginal portion of confirming current train image and reference picture, also comprises:
Utilize the statistics with histogram method that current train image and reference picture are carried out brightness normalization processing.
14. method as claimed in claim 12 is characterized in that, after the marginal portion of confirming current train image and reference picture, also comprises: the edge of current train image and reference picture is carried out the normalization processing.
15. a train abnormality detection system is characterized in that, comprising:
Image acquisition unit is used to obtain current train global image;
The train license number is confirmed the unit, is used for confirming current train license number;
Reference picture is chosen the unit, is used for selecting and the corresponding global image of current train license number image as a reference in preset image library;
The alignment comparing unit is used for the comparison of aliging with said reference picture of said current train global image;
The train abnormal area is confirmed the unit; Be used for alignment comparison result according to said alignment comparing unit; The inconsistent zone of confirming said current train global image and said reference picture is the train abnormal area; And, severely subnormal is carried out the classification of emphasis Fault Identification according to the order of severity of abnormal area.
16. system as claimed in claim 15 is characterized in that, also comprises:
The legend unit is used for said train abnormal area is carried out the legend sign, and perhaps, when having a plurality of abnormal area, the abnormal area that rank is surpassed preset thresholding carries out the legend sign.
17. like claim 15 or 16 described systems, it is characterized in that, also comprise:
Image deposits the unit in, is used for depositing said current train global image in said preset image library.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110155792.6A CN102323070B (en) | 2011-06-10 | 2011-06-10 | Method and system for detecting abnormality of train |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110155792.6A CN102323070B (en) | 2011-06-10 | 2011-06-10 | Method and system for detecting abnormality of train |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102323070A true CN102323070A (en) | 2012-01-18 |
| CN102323070B CN102323070B (en) | 2015-04-08 |
Family
ID=45450859
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110155792.6A Active CN102323070B (en) | 2011-06-10 | 2011-06-10 | Method and system for detecting abnormality of train |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102323070B (en) |
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103077526A (en) * | 2013-02-01 | 2013-05-01 | 苏州华兴致远电子科技有限公司 | Train abnormality detection method and system with deep detection function |
| CN103512762A (en) * | 2012-06-29 | 2014-01-15 | 北京华兴致远科技发展有限公司 | Image processing method and device and train fault detection system |
| CN104200464A (en) * | 2014-08-19 | 2014-12-10 | 苏州华兴致远电子科技有限公司 | Train anomaly recognition detection method and system |
| CN104463235A (en) * | 2014-11-18 | 2015-03-25 | 中国铁道科学研究院电子计算技术研究所 | Fault recognition method and device based on operation images of motor train unit |
| CN104634592A (en) * | 2014-12-18 | 2015-05-20 | 成都铁安科技有限责任公司 | Train running gear fault diagnosis method and train running gear fault diagnosis device |
| CN104809427A (en) * | 2014-12-31 | 2015-07-29 | 朱金良 | 3D information based fault automatic identification and detection system and detection method |
| CN105109521A (en) * | 2015-08-20 | 2015-12-02 | 苏州华兴致远电子科技有限公司 | Rail train anomaly detection method and system |
| CN106226050A (en) * | 2016-07-15 | 2016-12-14 | 北京航空航天大学 | A kind of TFDS fault automatic identifying method |
| CN106254839A (en) * | 2016-09-30 | 2016-12-21 | 湖南中铁五新重工有限公司 | The anti-method and device of slinging of container truck |
| CN106548147A (en) * | 2016-11-02 | 2017-03-29 | 南京鑫和汇通电子科技有限公司 | A kind of quick noise robustness image foreign matter detection method and TEDS systems |
| CN107220996A (en) * | 2017-05-17 | 2017-09-29 | 武汉大学 | A kind of unmanned plane linear array consistent based on three-legged structure and face battle array image matching method |
| CN108225706A (en) * | 2016-12-15 | 2018-06-29 | 北京唐智科技发展有限公司 | A kind of automated diagnostic method for identifying the intensive stripping pitted skin failure of wheel |
| CN108510019A (en) * | 2017-02-28 | 2018-09-07 | 北京天乐泰力科技发展有限公司 | A kind of intelligent platform systems and its control method based on RFID |
| CN108960344A (en) * | 2018-08-03 | 2018-12-07 | 敦煌研究院 | Difference detecting method, device and the terminal device of cultural relic images |
| CN109754410A (en) * | 2019-01-03 | 2019-05-14 | 北京化工大学 | A kind of railway car method of counting based on machine vision |
| CN110017998A (en) * | 2019-03-26 | 2019-07-16 | 中国人民财产保险股份有限公司 | Vehicle checking method, device and equipment |
| CN110217265A (en) * | 2019-06-21 | 2019-09-10 | 中国神华能源股份有限公司 | Vehicle flat sliding detection system and method |
| CN111902830A (en) * | 2018-03-23 | 2020-11-06 | 住友重机械工业株式会社 | Support device for construction machine |
| CN113781418A (en) * | 2021-08-30 | 2021-12-10 | 大连地铁集团有限公司 | Subway image anomaly detection method and system based on comparison and storage medium |
| WO2022007118A1 (en) * | 2020-07-06 | 2022-01-13 | 周爱丽 | Seat arrangement uniformity measurement system and method |
| CN116309741A (en) * | 2023-05-22 | 2023-06-23 | 中南大学 | TVDS image registration method, segmentation method, device and medium |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1996194A (en) * | 2005-12-31 | 2007-07-11 | 清华大学 | Moving body positioning and rectifying system and its motion tracking method |
| CN101334266A (en) * | 2008-07-18 | 2008-12-31 | 北京优纳科技有限公司 | Circuit board defect off-line checking method based on large-capacity image storage technology |
| US20090309966A1 (en) * | 2008-06-16 | 2009-12-17 | Chao-Ho Chen | Method of detecting moving objects |
| CN101616310A (en) * | 2009-07-17 | 2009-12-30 | 清华大学 | The target image stabilizing method of binocular vision system of variable visual angle and resolution |
| US20100131442A1 (en) * | 2008-11-21 | 2010-05-27 | Ming-Hwa Sheu | Intelligent monitoring system |
| CN101739829A (en) * | 2009-12-03 | 2010-06-16 | 北京中星微电子有限公司 | Video-based vehicle overspeed monitoring method and system |
| CN101794439A (en) * | 2010-03-04 | 2010-08-04 | 哈尔滨工程大学 | Image splicing method based on edge classification information |
| CN101957325A (en) * | 2010-10-14 | 2011-01-26 | 山东鲁能智能技术有限公司 | Substation equipment appearance abnormality recognition method based on substation inspection robot |
-
2011
- 2011-06-10 CN CN201110155792.6A patent/CN102323070B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1996194A (en) * | 2005-12-31 | 2007-07-11 | 清华大学 | Moving body positioning and rectifying system and its motion tracking method |
| US20090309966A1 (en) * | 2008-06-16 | 2009-12-17 | Chao-Ho Chen | Method of detecting moving objects |
| CN101334266A (en) * | 2008-07-18 | 2008-12-31 | 北京优纳科技有限公司 | Circuit board defect off-line checking method based on large-capacity image storage technology |
| US20100131442A1 (en) * | 2008-11-21 | 2010-05-27 | Ming-Hwa Sheu | Intelligent monitoring system |
| CN101616310A (en) * | 2009-07-17 | 2009-12-30 | 清华大学 | The target image stabilizing method of binocular vision system of variable visual angle and resolution |
| CN101739829A (en) * | 2009-12-03 | 2010-06-16 | 北京中星微电子有限公司 | Video-based vehicle overspeed monitoring method and system |
| CN101794439A (en) * | 2010-03-04 | 2010-08-04 | 哈尔滨工程大学 | Image splicing method based on edge classification information |
| CN101957325A (en) * | 2010-10-14 | 2011-01-26 | 山东鲁能智能技术有限公司 | Substation equipment appearance abnormality recognition method based on substation inspection robot |
Non-Patent Citations (1)
| Title |
|---|
| 刘化冰等: "GPS/DR/GIS技术在基于GSM-R列车监控***中应用", 《铁路航测》, no. 1, 30 March 2003 (2003-03-30) * |
Cited By (34)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103512762B (en) * | 2012-06-29 | 2016-12-21 | 北京华兴致远科技发展有限公司 | Image processing method, device and train failure detection system |
| CN103512762A (en) * | 2012-06-29 | 2014-01-15 | 北京华兴致远科技发展有限公司 | Image processing method and device and train fault detection system |
| CN103077526A (en) * | 2013-02-01 | 2013-05-01 | 苏州华兴致远电子科技有限公司 | Train abnormality detection method and system with deep detection function |
| CN104200464A (en) * | 2014-08-19 | 2014-12-10 | 苏州华兴致远电子科技有限公司 | Train anomaly recognition detection method and system |
| CN104200464B (en) * | 2014-08-19 | 2018-03-09 | 苏州华兴致远电子科技有限公司 | A kind of detection method and system for identifying that train is abnormal |
| CN104463235B (en) * | 2014-11-18 | 2019-01-18 | 中国铁道科学研究院电子计算技术研究所 | Fault recognition method and device based on EMU operation image |
| CN104463235A (en) * | 2014-11-18 | 2015-03-25 | 中国铁道科学研究院电子计算技术研究所 | Fault recognition method and device based on operation images of motor train unit |
| CN104634592A (en) * | 2014-12-18 | 2015-05-20 | 成都铁安科技有限责任公司 | Train running gear fault diagnosis method and train running gear fault diagnosis device |
| CN104809427A (en) * | 2014-12-31 | 2015-07-29 | 朱金良 | 3D information based fault automatic identification and detection system and detection method |
| CN104809427B (en) * | 2014-12-31 | 2019-02-15 | 朱金良 | Failure automatic identification detection system and detection method based on 3D information |
| CN105109521A (en) * | 2015-08-20 | 2015-12-02 | 苏州华兴致远电子科技有限公司 | Rail train anomaly detection method and system |
| CN105109521B (en) * | 2015-08-20 | 2017-07-07 | 苏州华兴致远电子科技有限公司 | A kind of track train method for detecting abnormality and system |
| CN106226050A (en) * | 2016-07-15 | 2016-12-14 | 北京航空航天大学 | A kind of TFDS fault automatic identifying method |
| CN106226050B (en) * | 2016-07-15 | 2019-02-01 | 北京航空航天大学 | A kind of TFDS fault picture automatic identifying method based on convolutional neural networks |
| CN106254839A (en) * | 2016-09-30 | 2016-12-21 | 湖南中铁五新重工有限公司 | The anti-method and device of slinging of container truck |
| CN106548147A (en) * | 2016-11-02 | 2017-03-29 | 南京鑫和汇通电子科技有限公司 | A kind of quick noise robustness image foreign matter detection method and TEDS systems |
| CN106548147B (en) * | 2016-11-02 | 2019-07-12 | 南京鑫和汇通电子科技有限公司 | A kind of quick noise robustness image foreign matter detection method and TEDS system |
| CN108225706B (en) * | 2016-12-15 | 2019-09-10 | 北京唐智科技发展有限公司 | A kind of identification wheel intensively removes the automated diagnostic method of pitted skin failure |
| CN108225706A (en) * | 2016-12-15 | 2018-06-29 | 北京唐智科技发展有限公司 | A kind of automated diagnostic method for identifying the intensive stripping pitted skin failure of wheel |
| CN108510019A (en) * | 2017-02-28 | 2018-09-07 | 北京天乐泰力科技发展有限公司 | A kind of intelligent platform systems and its control method based on RFID |
| CN108510019B (en) * | 2017-02-28 | 2021-02-05 | 北京天乐泰力科技发展有限公司 | Intelligent platform system based on RFID and control method thereof |
| CN107220996A (en) * | 2017-05-17 | 2017-09-29 | 武汉大学 | A kind of unmanned plane linear array consistent based on three-legged structure and face battle array image matching method |
| CN107220996B (en) * | 2017-05-17 | 2019-07-09 | 武汉大学 | One kind is based on the consistent unmanned plane linear array of three-legged structure and face battle array image matching method |
| CN111902830A (en) * | 2018-03-23 | 2020-11-06 | 住友重机械工业株式会社 | Support device for construction machine |
| CN108960344A (en) * | 2018-08-03 | 2018-12-07 | 敦煌研究院 | Difference detecting method, device and the terminal device of cultural relic images |
| CN109754410B (en) * | 2019-01-03 | 2020-12-11 | 北京化工大学 | Railway vehicle carriage counting method based on machine vision |
| CN109754410A (en) * | 2019-01-03 | 2019-05-14 | 北京化工大学 | A kind of railway car method of counting based on machine vision |
| CN110017998A (en) * | 2019-03-26 | 2019-07-16 | 中国人民财产保险股份有限公司 | Vehicle checking method, device and equipment |
| CN110017998B (en) * | 2019-03-26 | 2021-07-06 | 中国人民财产保险股份有限公司 | Vehicle detection method, device and equipment |
| CN110217265A (en) * | 2019-06-21 | 2019-09-10 | 中国神华能源股份有限公司 | Vehicle flat sliding detection system and method |
| WO2022007118A1 (en) * | 2020-07-06 | 2022-01-13 | 周爱丽 | Seat arrangement uniformity measurement system and method |
| CN113781418A (en) * | 2021-08-30 | 2021-12-10 | 大连地铁集团有限公司 | Subway image anomaly detection method and system based on comparison and storage medium |
| CN116309741A (en) * | 2023-05-22 | 2023-06-23 | 中南大学 | TVDS image registration method, segmentation method, device and medium |
| CN116309741B (en) * | 2023-05-22 | 2023-08-11 | 中南大学 | TVDS image registration method, segmentation method, device and medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102323070B (en) | 2015-04-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102323070A (en) | Method and system for detecting abnormality of train | |
| CN103077526B (en) | There is train method for detecting abnormality and the system of depth detection function | |
| CN109949578B (en) | Vehicle line pressing violation automatic auditing method based on deep learning | |
| CN103559703B (en) | Crane barrier based on binocular vision is monitored and method for early warning and system | |
| CN104029680B (en) | Lane Departure Warning System based on monocular cam and method | |
| CN106548147B (en) | A kind of quick noise robustness image foreign matter detection method and TEDS system | |
| CN103512762B (en) | Image processing method, device and train failure detection system | |
| CN105243381B (en) | Failure automatic identification detection system and method based on 3D information | |
| CN104392239B (en) | A kind of licence plate recognition method and system | |
| CN105426864A (en) | Multiple lane line detecting method based on isometric peripheral point matching | |
| CN106169244A (en) | The guidance information utilizing crossing recognition result provides device and method | |
| CN106461400A (en) | Vehicle positioning or navigation utilizing associated feature pairs | |
| CN112329747B (en) | Vehicle parameter detection method based on video identification and deep learning and related device | |
| CN102542843A (en) | Early warning method for preventing vehicle collision and device | |
| CN104809427B (en) | Failure automatic identification detection system and detection method based on 3D information | |
| JP2010205160A (en) | Method for notifying speed-limit sign recognition result | |
| CN112883948B (en) | Semantic segmentation and edge detection model building and guardrail abnormity monitoring method | |
| CN109460787A (en) | IDS Framework method for building up, device and data processing equipment | |
| CN101369312B (en) | Method and equipment for detecting intersection in image | |
| CN112418213A (en) | Vehicle driving track identification method and device and storage medium | |
| CN112488995A (en) | Intelligent injury judging method and system for automatic train maintenance | |
| CN105070094A (en) | Machine vision based parking space detection system and parking space detection method | |
| CN113085893A (en) | Vehicle control method and device based on image recognition | |
| CN103661491A (en) | System and method for intelligently controlling platform shielding gate | |
| KR101736431B1 (en) | System and Method for Smart Traffic Monitoring Using Multiple Image |
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 | ||
| C41 | Transfer of patent application or patent right or utility model | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20160426 Address after: Dongping street Suzhou City Industrial Park in Jiangsu province 215123 No. 270 building 4A, Australia Shunchang 4C unit Patentee after: Suzhou New Vision Science and Technology Co., Ltd. Address before: 100193, 107-A, 10 Zhongguancun Software Park, 8 northeast Wang Xi Road, Beijing, Haidian District Patentee before: Beijing Huaxing Zhiyuan Technology Development Co., Ltd. |