CN105550654B - Bullet train image capturing system, real-time license number detection system and method - Google Patents
Bullet train image capturing system, real-time license number detection system and method Download PDFInfo
- Publication number
- CN105550654B CN105550654B CN201510938317.4A CN201510938317A CN105550654B CN 105550654 B CN105550654 B CN 105550654B CN 201510938317 A CN201510938317 A CN 201510938317A CN 105550654 B CN105550654 B CN 105550654B
- Authority
- CN
- China
- Prior art keywords
- license number
- train
- image
- module
- time
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
- G06V20/584—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads of vehicle lights or traffic lights
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/14—Image acquisition
- G06V30/146—Aligning or centring of the image pick-up or image-field
- G06V30/1475—Inclination or skew detection or correction of characters or of image to be recognised
- G06V30/1478—Inclination or skew detection or correction of characters or of image to be recognised of characters or characters lines
Abstract
The invention discloses a kind of bullet train image capturing system, real-time license number detection system and methods, image capturing system includes train speed detection module, Image Acquisition control module, license number image capture module, train speed detection module is used to detect arrival time T and train speed v that train reaches designated position, and Image Acquisition control module is used to control license number image capture module at the appointed time T according to arrival time T and train speed vsStart to acquire train image, and later according to certain time interval TiAcquire train image.The present invention can automatically adjust Image Acquisition interval time according to train length in train running speed, imaging device visual field, preferably to obtain train license number image, in order to which the identification of subsequent license number can be more accurate and high-quality, to the license number of bullet train real-time and accurately automatic identification.
Description
Technical field
The present invention relates to train image processing technology fields, and in particular, to a kind of bullet train image capturing system, reality
When license number detection system and method.
Background technique
With the continuous development of China express railway, while overall trip speed is continuously improved in train, the peace of train operation
Full property problem also becomes more and more important, and is related to the safety of life and property of the people.For this purpose, guaranteeing that train operation is safely
The core of all working.As China's Urbanization Progress deepens continuously, the mobility between region is dramatically increased, to bullet train
Demand be also continuously increased, the density and duration of train operation are also increase accordingly.Therefore, only by train detection personnel to train
The time detected also accordingly is reduced, it is often more important that and the safety problem of train during operation cannot be timely responded to,
To reduce the traffic safety of bullet train.The security information of train, such as pantograph are obtained in the train operation phase in order to realize
Working condition etc., then can use the mode that a certain position of train is imaged and running train be monitored.So
And while acquisition train a certain site image information, it is also necessary to which the license number for obtaining the train could be by image information and column
Vehicle is corresponded to, in order to check the rapid orientation problem train of service personnel.Therefore, column are obtained in bullet train operational process
Vehicle license number is particularly important.
The implementation of existing train car number automatic identification system mainly installs electronic tag ON TRAINS, all
District station, marshalling yard and divisional station installation ground identification equipment (AEI), when train passes through, AEI equipment is by reading electronics mark
Label obtain train license number.The system needs to install more electronic equipment ON TRAINS, including microstrip antenna, virtual power supply, anti-
Penetrate modulator, encoder, microprocessor and memory etc., the heavy workload that equipment is installed, safeguarded, if electronic tag is damaged
Bad, AEI equipment will be unable to obtain the case where vehicle number information leads to erroneous judgement and leaks vehicle.
With the continuous development of image recognition technology, there are also realize train Train number recognition using image recognition technology
System, the problem of being primarily present, have: can not achieve the Train number recognition of bullet train (being greater than 200km/h), the standard of Train number recognition
Not high, train license number poor image quality of night acquisition of true rate etc..
Summary of the invention
The object of the invention is that the shortcomings that overcoming the above-mentioned prior art and deficiency, a kind of bullet train image is provided and is adopted
Collecting system, the system can automatically adjust Image Acquisition interval according to train length in train running speed, imaging device visual field
Time, preferably to obtain train license number image, in order to which the identification of subsequent license number can be more accurate and high-quality;The present invention is also
The real-time license number detection system of bullet train and the real-time license number detection method of bullet train are provided, it can be to the license number of bullet train
Real-time and accurately automatic identification.
Technical solution used by the present invention solves the above problems is:
Bullet train image capturing system includes that train speed detection module, Image Acquisition control module, license number image are adopted
Collecting module, in which: train speed detection module reaches the arrival time T and train speed v of designated position for detecting train,
And the arrival time T and train speed v that will test are sent to Image Acquisition control module;License number image capture module, is used for
Acquire train image;Image Acquisition control module, for controlling license number Image Acquisition mould according to arrival time T and train speed v
Block at the appointed time TsStart to acquire train image, and later according to certain time interval TiAcquire train image.
As a further improvement of the present invention, the specified time TsCalculation be: Ts=(d-v*Tf)/v+T+Tf
(1);
In formula (1), TfDuration is filtered for the anti-error bat of license number image capture module, d detects mould for designated position to train speed
The distance of block.
As the another improvement time interval T of the inventioniValue are as follows: Ti≦w/v (2);
In formula (2), w is the visual field width of license number image capture module.In the technical program, can according to train running speed,
The visual field width of image capture module automatic adjustment Image Acquisition interval time (directly proportional to train length in visual field), with more preferable
Ground obtains train license number image, real to the license number of bullet train in order to which the identification of subsequent license number can be more accurate and high-quality
When accurately automatic identification.
Further, the license number image capture module is opened with the picture acquired every time not less than N, wherein N=a/w,
In, a is train rail cars length.
Further, the time for exposure of the license number image capture module is Tp, TpCalculation be: Tp=s/v (3);
In formula (3), s be the time for exposure in allow license number move maximum distance, s=r*n/180 (4),
Wherein, n is that license number image capture module allows license number to move within the time for exposure in the case where guaranteeing image clearly
Maximum central angle, r are distance of the license number image capture module to train license number.It, can be according to train operation in the technical program
Speed, the parameter of Image Acquisition control module adjust the time for exposure, preferably to obtain train license number image, guarantee train high speed
Image Acquisition quality under operating status, in order to which the identification of subsequent license number can be more accurate and high-quality.
Preferably, the camera lens shooting direction of the license number image capture module and train are at 30-60 ° of angle.This technology side
In case, shooting direction and the train of Image Acquisition control module form an angle, and the shooting for being conducive to increase train in image is long
Degree, utmostly shoots license number into picture, convenient for the identification of license number when image procossing.
The real-time license number detection system of bullet train, including image collecting device and image analysis apparatus, described image acquisition
Device is the bullet train image capturing system in any of the above-described technical solution, and described image analytical equipment includes image procossing mould
Block and license number statistical module,
Described image processing module includes: license number locating module, for obtaining train license number region in image;License number perspective
Conversion module, for making in train license number region license number become horizontality from heeling condition;License number character extraction module, is used for
Piecemeal segmentation is carried out to image, extracts the character in license number;Train number recognition, the word that license number character extraction module extracts for identification
Symbol obtains the train license number in image;
The license number statistical module is used to count the selected relatively position in all license numbers identified, by selected relatively position
The corresponding license number of the identical most image of character is set as correct license number.
The real-time license number detection method of bullet train, including image acquisition step and image analysis step, which is characterized in that institute
State image acquisition step are as follows: image collecting device at the appointed time TsStart to acquire train image, and later according to certain
Time interval TiAcquire train image;
Described image analytical procedure includes:
License number in S1, all images of identification identifies the license number in every image in the step;
Selected relatively position in S2, all license numbers identified of statistics, the character of selected relatively position is identical most
The corresponding license number of more images is set as correct license number;
The step of identifying in step S1 to the license number in every image is as follows:
S11, license number positioning: train license number region in image is obtained;
S12, license number character extract: carrying out piecemeal segmentation to image, extract the character in license number;
S13, Train number recognition: the character extracted in identification step S12 obtains the train license number in image.
In the technical program, the image acquired in license number detection method is train license number sequence image, by sequence chart
As carrying out Train number recognition, the accuracy of Train number recognition can be improved.
Further, also there is license number perspective transform step between step S11 and step S12, make in train license number region
License number becomes horizontality from heeling condition, license number perspective transform method particularly includes: enables the train license number obtained in step S11
The coordinate of four angle points in region is respectively as follows: (x0, y0), (x1, y1), (x2, y2), (x3, y3);Using Shear Transform by this 4
The coordinate pair of a angle point should be transformed to (0,0), (x, 0), (x, y), (0, y).In the technical program, used in license number detection method
Train license number transformation based on perspective transform, can be improved Train number recognition accuracy.
Further, specified time TsCalculation method are as follows: Ts=(d-v*Tf)/v+T+Tf,
Time interval TiCalculation method are as follows: Ti=w/v;
Wherein, the arrival time when train arrival designated position that T detects for train speed detection module;TfIt is anti-error
The filtering duration of bat, v are the train speed that train speed detection module detects, d detects mould for designated position to train speed
The distance of block, w are the visual field width of license number image capture module.
To sum up, the beneficial effects of the present invention are:
1, the present invention can be according between train length's automatic adjustment Image Acquisition in train running speed, imaging device visual field
Every the time, preferably to obtain train license number image, in order to which the identification of subsequent license number can be more accurate and high-quality, to high speed
The license number of train real-time and accurately automatic identification;
2, the present invention can adjust the time for exposure according to the parameter of train running speed, Image Acquisition control module, with more
Train license number image is obtained well, guarantees the Image Acquisition quality under train high-speed cruising state, in order to the knowledge of subsequent license number
It can not be more accurate and high-quality;
3, the shooting direction of Image Acquisition control module of the invention and train form an angle, and are conducive to increase in image
The shooting length of train utmostly shoots license number into picture, convenient for the identification of license number when image procossing;
4, the image acquired in the license number detection method in the present invention is train license number sequence image, by sequence image
Train number recognition is carried out, can be improved the accuracy of Train number recognition;
5, license number can be improved using the train license number transformation based on perspective transform in the license number detection method in the present invention
Identify accuracy.
Detailed description of the invention
Fig. 1 is the visual field schematic diagram of license number image capture module of the invention;
Fig. 2 is the schematic diagram of present invention Shear Transform in image procossing.
Specific embodiment
It can not achieve the license number of bullet train (being greater than 200km/h) to solve train car number identification system in the prior art
The problems such as identification, the train license number poor image quality that the accuracy rate of Train number recognition is high, night obtains, the present invention provides a kind of high
Fast train image acquisition system, the system can be automatically adjusted according to train length in train running speed, imaging device visual field
Image Acquisition interval time, preferably to obtain train license number image, in order to subsequent license number identification can it is more accurate and
It is high-quality;The present invention also provides the real-time license number detection systems of bullet train and the real-time license number detection method of bullet train, can be right
200km/h or more speed per hour even the license number real-time and accurately automatic identification of the bullet train of 350km/h.
The present invention will now be described in further detail with reference to the accompanying drawings and the accompanying drawings, but embodiments of the present invention
It is without being limited thereto.
Embodiment 1:
Bullet train image capturing system, including train speed detection module, Image Acquisition control module, license number image are adopted
Collect module, in which:
Train speed detection module reaches the arrival time T and train speed v of designated position for detecting train, and will
The arrival time T and train speed v detected is sent to Image Acquisition control module;
License number image capture module uses camera in the present embodiment for acquiring train image;
Image Acquisition control module, for being referred to according to arrival time T and train speed v control license number image capture module
Fix time TsStart to acquire train image, and later according to certain time interval TiAcquire train image.
Train speed detection module mainly detects train speed by velocity radar, can be judged according to radar data
The arrival of train and its real-time speed, the maximum distance of radar detection on the direction that above-mentioned designated position, that is, train is arrived.
Train speed detection module is the preposition trigger device of entire bullet train image capturing system, and detection accuracy determines image
The interval time at the time point of acquisition and every frame image.
If the maximum distance of radar detection is d, the time (i.e. arrival time) of radar detection to train is T, train speed
For v;The visual field width of camera is w, when filtering of anti-error bat a length of Tf, then:
Start time (the i.e. specified time T of triggering camera collection images) calculation be:
Ts=(d-v*Tf)/v+T+Tf(1);
The time interval T of two field pictures shootingiValue are as follows: Ti≦w/v (2);
T is just taken in the present embodimenti=w/v.W/v is time interval TiMaximum value, the maximum time interval master of two field pictures
If in order to guarantee that the visual field of adjacent two field pictures has certain coincidence, to guarantee to collect license number image.
The license number image capture module is with the picture acquired every time not less than N, wherein N=a/w, wherein a is column
Vehicle rail cars length.
Above-mentioned camera visual field width w is directly proportional to train length L in camera visual field, train in visual field width and visual field
The signal of length as shown in Figure 1, in Fig. 1 the intersection point of dotted line x and train to the horizontal distance between dotted line y and the intersection point of train just
It is visual field width, the distance between intersection point of dotted line x and train to the intersection point of dotted line y and train is train length in market.
In the present embodiment, it can be adjusted automatically according to train length in train running speed, imaging device (camera) visual field
Image Acquisition interval time is saved, preferably to obtain train license number image, in order to which the identification of subsequent license number can be more accurate
With it is high-quality.
Embodiment 2:
On the basis of embodiment 1, license number image-forming module is improved as follows in the present embodiment:
In the present embodiment, Image Acquisition control module should use technical grade high-speed imaging equipment, in the way of external trigger
Train license number sequence image is obtained, in order to guarantee the Image Acquisition quality under train high-speed cruising state, the exposure of imaging device
Time must satisfy the needs of high speed acquisition, if imaging device allows object in the time for exposure in the case where guaranteeing image clearly
The maximum central angle of interior movement is n, and the distance of imaging device to train license number is r, then:
The maximum distance s for allowing license number to move in time for exposure are as follows: s=r*n/180 (4);
To which the time for exposure T of the license number image capture module can be obtainedpAre as follows: Tp=s/v (3).
In the present embodiment, the time for exposure can be adjusted according to the parameter of train running speed, Image Acquisition control module, with
Train license number image is preferably obtained, guarantees the Image Acquisition quality under train high-speed cruising state, in order to subsequent license number
Identification can be more accurate and high-quality.
Embodiment 3:
On the basis of embodiment 1 or embodiment 2, license number image capture module is improved as follows in the present embodiment:
The visual field width of known image acquisition control module is w, in order to increase the length of car body included in single image
It spends L and collects train license number in favor of better ensuring that:
As shown in Figure 1, the camera lens shooting direction and train of the license number image capture module are at 30-60 ° of angle, i.e. Fig. 1
In ∠ Φ be 30-60 °.Preferably, in the present embodiment, the camera lens shooting direction of the license number image capture module and train at
45 ° of angles, i.e. Φ=45 ° ∠.In order to collect train license number in train high-speed cruising, the camera lens of imaging device in this system
Shooting direction (direction where straight line z) and train angle at 45 °, rather than vertical train shooting, are conducive to increase image in this way
The shooting length of middle train utmostly shoots license number into picture, convenient for the identification of license number when image procossing, if ∠ Φ is too
Small, then the number plate region shot is unclear, therefore is not more than 60 degree, and optimal selection is 45 degree.
Embodiment 4:
On the basis of embodiment 1- embodiment 3 is any, bullet train image capturing system is carried out such as in the present embodiment
Lower improvement:
Bullet train image capturing system further includes light compensating module, and the light compensating module is using infrared light triggering frequency
Sudden strain of a muscle light filling mode is license number image capture module light filling.Bullet train image capturing system is beside railroad track, if adopted
With visible light source, then traffic safety can be impacted.Simultaneously as the time for exposure of imaging device setting is shorter, in order to
Guarantee the brightness of acquired image, light compensating module all mends all weather operations, i.e. day and night to imaging device
Light.
Embodiment 5:
The real-time license number detection system of bullet train, including image collecting device and image analysis apparatus, described image acquisition
Device is any bullet train image capturing system in embodiment 1-4, and described image analytical equipment includes image procossing
Module and license number statistical module,
Described image processing module includes:
License number locating module, for obtaining train license number region in image;
License number perspective transform module, for making in train license number region license number become horizontality from heeling condition;
License number character extraction module extracts the character in license number for carrying out piecemeal segmentation to image;
Train number recognition, the character that license number character extraction module extracts for identification obtain the train license number in image;
The license number statistical module is used to count the selected relatively position in all license numbers identified, by selected relatively position
The corresponding license number of the identical most image of character is set as correct license number.
Embodiment 6:
The real-time license number detection method of bullet train is carried out using the real-time license number detection system of bullet train in embodiment 5
Detection, including image acquisition step and image analysis step.
Described image acquisition step are as follows: image collecting device at the appointed time TsStart to acquire train image, and later
According to certain time interval TiAcquire train image, specified time TsCalculation method are as follows: Ts=(d-v*Tf)/v+T+Tf, when
Between be spaced TiCalculation method are as follows: Ti=w/v;Wherein, T is that the train that train speed detection module detects reaches designated position
When arrival time;TfFor the filtering duration of anti-error bat, v is the train speed that train speed detection module detects, d is specified
To the distance of train speed detection module, w is the visual field width of license number image capture module for position.
Described image analytical procedure, which is realized, carries out Train number recognition function to collected train license number image, and main includes pair
Each image in sequence image carries out Train number recognition and carries out comprehensive analysis to sequence image Train number recognition result to obtain
Correct license number and corresponding image include step S1 and S2:
License number in S1, all images of identification identifies the license number in every image in the step;
Selected relatively position in S2, all license numbers identified of statistics, the character of selected relatively position is identical most
The corresponding license number of more images is set as correct license number.
The each image train Train number recognition algorithm being directed in sequence image in step S1 mainly includes license number positioning, license number
Extraction, the SVM classifier study based on character HOG feature carry out Train number recognition etc. using classifier, specifically, scheme to every
The step of license number as in is identified is as follows:
S11, license number positioning: train license number region in image is obtained;According to feature of image, Hough transformation detection image is utilized
In straight line, wherein the minimum length threshold for needing to be arranged detected straight line is minLen, maximum linear gap threshold is
Then maxGap obtains license number region masking-out according to the straight line detected, carry out binaryzation to the masked area in original image
Obtain train license number region;
S12, license number character extract: the combination of multiple characters or letter when due in license number region carries out image
Piecemeal segmentation, extracts each character in license number;
S13, Train number recognition: calculating HOG characteristic value to each character for extracting in step S12, then input classifier into
Row identification, obtains the train license number in image.
Known by the license number that single image train Train number recognition algorithm has calculated each image in train license number sequence chart
S2 is entered step after other result.It is not that each image has train license number, while cannot guarantee that and accurately identify often in sequence chart
One width contains the image of license number, it is therefore desirable to analysis comparison is carried out to all results, to obtain correct recognition result.
For train license number, the license number of each compartment has certain rule.For example, same train, first four of license number
For license number, latter two are coach number, in the present embodiment, are set as selected relatively position for first four, count all vehicles identified
First four in number, finally determine that the identical amount of images of preceding four characters is at most correct license number.
Embodiment 7:
On the basis of embodiment 6, further, if the visual angle of imaging device be it is at a certain angle with train, therefore
License number is inclined in the license number image collected.Discrimination is extracted and improved in order to facilitate license number character, is needed to license number
Carry out perspective transform:
Also there is license number perspective transform step between step S11 and step S12, make in train license number region license number by inclining
Ramp-like state becomes horizontality, license number perspective transform method particularly includes: as shown in Fig. 2, enabling the train vehicle obtained in step S11
The coordinate of four angle points in number region is respectively as follows: the lower left corner (x0, y0), the lower right corner (x1, y1), the upper left corner (x2, y2), the upper right corner
(x3,y3);The coordinate pair of this 4 angle points should be transformed to by (0,0), (x, 0), (x, y), (0, y) using Shear Transform.
Enable Shear Transform formula are as follows:
The value of b and d should meet:
Above-mentioned equation group is solved to obtain:
When b and d take above-mentioned value, (x0, y0), (x1, y1), (x2, y2), (x3, y3) are substituted into Shear Transform formula and obtained
To new coordinate (0,0), (x, 0), (x, y), (0, y), using the train license number region after perspective transform as shown in Fig. 2, becoming
Horizontality, the lower left corner and the upper left corner are located in y-axis, and the lower left corner and the lower right corner are located in x-axis, subsequently into step S12 into
Sign character of driving a vehicle extracts, and easily facilitates the operations such as segmentation when character extracts.
Above-mentioned license number positions, character extracts, character recognition is all the common technology of technical field of image processing, this implementation
Its specific implementation is repeated no more in example.
The above is only the preferred embodiment of the present invention, protection scope of the present invention is not limited merely to above-described embodiment,
All technical solutions belonged under thinking of the present invention all belong to the scope of protection of the present invention.It should be pointed out that for the art
For those of ordinary skill, several improvements and modifications without departing from the principles of the present invention should be regarded as protection of the invention
Range.
Claims (7)
1. bullet train image capturing system, which is characterized in that including train speed detection module, Image Acquisition control module,
License number image capture module, in which:
Train speed detection module reaches the arrival time T and train speed v of designated position for detecting train, and will test
To arrival time T and train speed v be sent to Image Acquisition control module;
License number image capture module, for acquiring train image;
Image Acquisition control module, for controlling license number image capture module when specified according to arrival time T and train speed v
Between TsStart to acquire train image, and later according to certain time interval TiAcquire train image;
The specified time TsCalculation be: Ts=d/v+T (1);
In formula (1), d is distance of the designated position to train speed detection module;
The camera lens shooting direction and train of the license number image capture module are at 30-60 ° of angle;
The time for exposure of the license number image capture module is Tp, TpCalculation be: Tp=s/v (3);
In formula (3), s be the time for exposure in allow license number move maximum distance, s=r*n/180 (4),
Wherein, n is that license number image capture module allows license number to move within the time for exposure in the case where guaranteeing image clearly
Maximum central angle, r are distance of the license number image capture module to train license number.
2. bullet train image capturing system according to claim 1, which is characterized in that the time interval TiValue
Are as follows: Ti≦w/v (2);
In formula (2), w is the visual field width of license number image capture module.
3. bullet train image capturing system according to claim 1, which is characterized in that the license number image capture module
It is opened with the picture acquired every time not less than N, wherein N=a/w, wherein a is train rail cars length, and w is that license number image is adopted
Collect the visual field width of module.
4. the real-time license number detection system of bullet train, including image collecting device and image analysis apparatus, which is characterized in that described
Image collector is set to bullet train image capturing system as claimed in any one of claims 1-3, described image analytical equipment packet
Image processing module and license number statistical module are included,
Described image processing module includes:
License number locating module, for obtaining train license number region in image;
License number perspective transform module, for making in train license number region license number become horizontality from heeling condition;
License number character extraction module extracts the character in license number for carrying out piecemeal segmentation to image;
Train number recognition, the character that license number character extraction module extracts for identification obtain the train license number in image;
The license number statistical module is used to count the selected relatively position in all license numbers identified, by the character of selected relatively position
The corresponding license number of identical most image is set as correct license number.
5. the real-time license number detection method of bullet train, including image acquisition step and image analysis step, which is characterized in that use
The real-time license number detection system of bullet train in claim 4 carries out Image Acquisition and image analysis, described image acquisition step
Are as follows: image collecting device at the appointed time TsStart to acquire train image, and later according to certain time interval TiAcquisition column
Vehicle image,
Described image analytical procedure includes:
License number in S1, all images of identification identifies the license number in every image in the step;
Selected relatively position in S2, all license numbers identified of statistics, the character of selected relatively position is identical most
The corresponding license number of image is set as correct license number;
The step of identifying in step S1 to the license number in every image is as follows:
S11, license number positioning: train license number region in image is obtained;
S12, license number character extract: carrying out piecemeal segmentation to image, extract the character in license number;
S13, Train number recognition: the character extracted in identification step S12 obtains the train license number in image.
6. the real-time license number detection method of bullet train according to claim 5, which is characterized in that in step S11 and step
Also there is license number perspective transform step between S12, license number is made in train license number region to become horizontality, license number from heeling condition
Perspective transform method particularly includes: enable the coordinate of four angle points in the train license number region obtained in step S11 be respectively as follows: (x0,
y0),(x1,y1),(x2,y2),(x3,y3);Using Shear Transform by the coordinate pair of this 4 angle points should be transformed to (0,0), (x,
0)、(x,y)、(0,y)。
7. the real-time license number detection method of bullet train according to claim 5 or 6, which is characterized in that specified time TsMeter
Calculation method are as follows:
Ts=d/v+T,
Time interval TiCalculation method are as follows: Ti=w/v;
Wherein, the arrival time when train arrival designated position that T detects for train speed detection module;V is train speed
The train speed that detection module detects, d are distance of the designated position to train speed detection module, and w is license number Image Acquisition
The visual field width of module.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510938317.4A CN105550654B (en) | 2015-12-15 | 2015-12-15 | Bullet train image capturing system, real-time license number detection system and method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510938317.4A CN105550654B (en) | 2015-12-15 | 2015-12-15 | Bullet train image capturing system, real-time license number detection system and method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105550654A CN105550654A (en) | 2016-05-04 |
CN105550654B true CN105550654B (en) | 2019-01-29 |
Family
ID=55829837
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510938317.4A Active CN105550654B (en) | 2015-12-15 | 2015-12-15 | Bullet train image capturing system, real-time license number detection system and method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105550654B (en) |
Families Citing this family (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109318717A (en) * | 2017-07-31 | 2019-02-12 | 天津市松正电动汽车技术股份有限公司 | A kind of double source trolleybus image recognition control system and method |
CN109241971A (en) * | 2018-07-18 | 2019-01-18 | 东莞市诺丽电子科技有限公司 | A kind of municipal rail train car number identification system |
CN109344829B (en) * | 2018-07-23 | 2021-05-18 | 广州科易光电技术有限公司 | Train number identification method and device for high-speed railway train |
CN111669540A (en) * | 2019-03-07 | 2020-09-15 | 上海思桂信息技术有限公司 | Monitoring system and method based on AI technology |
CN110378251B (en) * | 2019-06-28 | 2021-04-27 | 湖南华菱涟源钢铁有限公司 | Control method and device for weighing system of train weighing department and readable storage medium |
CN112866673B (en) * | 2021-01-12 | 2022-10-21 | 北京格灵深瞳信息技术股份有限公司 | Image acquisition equipment, camera control method and device and train image acquisition system |
CN113327426B (en) * | 2021-05-26 | 2022-09-09 | 国能朔黄铁路发展有限责任公司 | Vehicle type code identification method and device and vehicle number identification method and device |
CN113537219A (en) * | 2021-08-05 | 2021-10-22 | 上海画骨智能科技有限公司 | Railway freight car number identification system |
CN114827463B (en) * | 2022-04-19 | 2023-09-22 | 中车青岛四方车辆研究所有限公司 | Image acquisition control system and method |
CN116161081B (en) * | 2023-04-18 | 2023-07-07 | 中铁电气化局集团有限公司 | Train positioning system and method based on trackside power grid pole counting |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101424529A (en) * | 2007-11-02 | 2009-05-06 | 北京华旗资讯数码科技有限公司 | Train scanning system |
CN201846437U (en) * | 2010-11-02 | 2011-05-25 | 大连鑫奇辉科技有限公司 | High-definition video monitoring system for truck loading state |
CN103294997A (en) * | 2013-05-20 | 2013-09-11 | 北京康拓红外技术股份有限公司 | Train number identification method based on static images |
CN203537492U (en) * | 2013-10-23 | 2014-04-09 | 苏州华兴致远电子科技有限公司 | Non-shape-transformation train image acquisition apparatus |
CN104401369A (en) * | 2014-10-17 | 2015-03-11 | 苏州华兴致远电子科技有限公司 | Train monitoring system and train monitoring method |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TWI409718B (en) * | 2009-12-04 | 2013-09-21 | Huper Lab Co Ltd | Method of locating license plate of moving vehicle |
-
2015
- 2015-12-15 CN CN201510938317.4A patent/CN105550654B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101424529A (en) * | 2007-11-02 | 2009-05-06 | 北京华旗资讯数码科技有限公司 | Train scanning system |
CN201846437U (en) * | 2010-11-02 | 2011-05-25 | 大连鑫奇辉科技有限公司 | High-definition video monitoring system for truck loading state |
CN103294997A (en) * | 2013-05-20 | 2013-09-11 | 北京康拓红外技术股份有限公司 | Train number identification method based on static images |
CN203537492U (en) * | 2013-10-23 | 2014-04-09 | 苏州华兴致远电子科技有限公司 | Non-shape-transformation train image acquisition apparatus |
CN104401369A (en) * | 2014-10-17 | 2015-03-11 | 苏州华兴致远电子科技有限公司 | Train monitoring system and train monitoring method |
Non-Patent Citations (1)
Title |
---|
基于视频列车车号识别系统的研究;陈春雷 等;《兰州交通大学学报(自然科学版)》;20071231;第26卷(第6期);第99-102页 |
Also Published As
Publication number | Publication date |
---|---|
CN105550654A (en) | 2016-05-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105550654B (en) | Bullet train image capturing system, real-time license number detection system and method | |
CN102759347B (en) | Online in-process quality control device and method for high-speed rail contact networks and composed high-speed rail contact network detection system thereof | |
CN105702048B (en) | Highway front truck illegal road occupation identifying system based on automobile data recorder and method | |
CN205920509U (en) | Vehicle charging information acquisition device who removes | |
CN110136447A (en) | Lane change of driving a vehicle detects and method for distinguishing is known in illegal lane change | |
CN107301776A (en) | Track road conditions processing and dissemination method based on video detection technology | |
CN103065519A (en) | Detecting system of ramp fixed point stopping and starting and detecting method of ramp fixed point stopping and starting | |
CN108986472A (en) | One kind turns around vehicle monitoring method and device | |
CN102129775A (en) | Method and system for obtaining evidence by capturing vehicles at traffic crossing under panoramic video detection | |
CN103745224A (en) | Image-based railway contact net bird-nest abnormal condition detection method | |
CN103279756A (en) | Vehicle detecting analysis system and detecting analysis method thereof based on integrated classifier | |
CN101882374A (en) | System and method for analysis of comprehensive information on vehicle trip distribution | |
CN105185118A (en) | Vehicle peccancy detection device and method based on video detection | |
CN103164958B (en) | Method and system for vehicle monitoring | |
CN105701453B (en) | A kind of railway with obstacle identification system is with quarrel vehicle and its obstacle identification method | |
CN110021167A (en) | A kind of highway cloud inspection system and data processing method | |
CN111696364A (en) | Roadside parking management method | |
CN109359596A (en) | A kind of highway vehicle localization method fast and accurately | |
CN204825605U (en) | Discernment of road environment and detection device based on image | |
CN108846331A (en) | The video frequency identifying method whether a kind of EMU chassis screw fastener falls off | |
CN106205135B (en) | A kind of detection method, apparatus and system and a kind of ball machine of the vehicle behavior violating the regulations that turns around | |
CN110070729A (en) | It is a kind of that vehicle detecting system and method are stopped based on the separated of mist calculating | |
CN205608812U (en) | Public transport passenger flow measures detecting system based on face identification and position location | |
CN107697102A (en) | Information gathering, identification and the caution system of high-speed railway foreign body intrusion | |
CN110321982A (en) | A kind of rail traffic section volume of the flow of passengers real-time computing technique |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |