CN110210427A - A kind of shelter bridge working condition detection system and method based on image processing techniques - Google Patents
A kind of shelter bridge working condition detection system and method based on image processing techniques Download PDFInfo
- Publication number
- CN110210427A CN110210427A CN201910489381.7A CN201910489381A CN110210427A CN 110210427 A CN110210427 A CN 110210427A CN 201910489381 A CN201910489381 A CN 201910489381A CN 110210427 A CN110210427 A CN 110210427A
- Authority
- CN
- China
- Prior art keywords
- shelter bridge
- area
- image frame
- bridge
- sample
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
Abstract
The invention discloses a kind of shelter bridge working condition detection system and method based on image processing techniques, including video acquisition system, data collection system, convolutional neural networks system and shelter bridge working condition detection system, the video acquisition system, data collection system, convolutional neural networks system, shelter bridge working condition detection system successively communicates to connect, the data collection system includes data image frame electric wire division module and data image frame acquisitions module, the convolutional neural networks internal system has shelter bridge position sample database, the shelter bridge working condition detection system includes shelter bridge working condition judgment module；The video acquisition system includes several video cameras.The present invention judges the current working condition of shelter bridge, can obtain beginning and deadline that beginning and deadline and shelter bridge of the shelter bridge by machine are disembarked simultaneously by automatically analyzing to machine level ground monitor video signal.
Description
Technical field
The present invention relates to shelter bridge working condition detection technique field more particularly to a kind of shelter bridges based on image processing techniques
Working condition detection system and method.
Background technique
The position of shelter bridge is detected in the monitor video of Flying Area in Airport, it, can be in conjunction with preset decision logic
The beginning and completion of aircraft movement are left in the beginning and deadline, shelter bridge acted to shelter bridge in flight support process close to aircraft
Time is automatically extracted, so that airfield support personnel grasp the critical processes of flight guarantee.Airport relies primarily on shelter bridge at present
The signal of bridge loading system passback judge the job state of shelter bridge, but the new shelter bridge quantity of one side mounting bridge loading system compared with
Few, and improvement expenses are higher, on the other hand, even if mounting bridge loading system, a possibility that there is also equipment faults, video detection can
To play the role of data backup.
Summary of the invention
Place in view of the shortcomings of the prior art, the purpose of the present invention is to provide a kind of based on image processing techniques
Shelter bridge working condition detection system and method judge the current work of shelter bridge by automatically analyzing to machine level ground monitor video signal
Situation can obtain beginning and deadline that beginning and deadline and shelter bridge of the shelter bridge by machine are disembarked simultaneously.
The purpose of the invention is achieved by the following technical solution:
A kind of shelter bridge working condition detection system based on image processing techniques, including the acquisition of video acquisition system, data
System, convolutional neural networks system and shelter bridge working condition detection system, the video acquisition system, data collection system, volume
Product nerve network system, shelter bridge working condition detection system successively communicate to connect, and the data collection system includes data image
Frame electric wire division module and data image frame acquisitions module, the convolutional neural networks internal system have shelter bridge position sample
Database, the shelter bridge working condition detection system include shelter bridge working condition judgment module；The video acquisition system includes
Several video cameras, the video acquisition system is for acquiring video data and combination of transmitted to data collection system, institute
The data image frame acquisition module for stating data collection system is used to carry out video data video image frame acquisition one by one and will view
Frequency picture frame is transmitted to data image frame electric wire division module, and the data image frame electric wire division module is used for video
Picture frame delineates out electric wire A and video image frame is divided into the area A1 and the area A2, the electronics by electric wire B, the electric wire A
The A2 zoning of video image frame is divided into the area A2B1 and the area A2B2 by line B, and the area A2B1 is close to the area A1；The shelter bridge position sample data
Library is used to be stored with the sample data of shelter bridge working condition, and the sample data includes that shelter bridge starts by bridge sample data, shelter bridge
Initially moving off sample data and shelter bridge work by bridge completed sample evidence, shelter bridge terminates sample data；The convolutional neural networks
System be used for data collection system institute collected video image frame and shelter bridge position sample database sample data progress
Image characteristics extraction and comparison processing, the shelter bridge working condition judgment module are used for the sample according to shelter bridge position sample database
Notebook data show that the working condition of shelter bridge, the working condition of shelter bridge include by image characteristics extraction, comparison processing, logic judgment
Shelter bridge start by bridge, shelter bridge by bridge complete, shelter bridge initially move off with shelter bridge work terminate.
In order to which shelter bridge working condition detection system of the present invention is better achieved, the convolutional neural networks system is used for logarithm
Shelter bridge characteristic area and background characteristics region, the convolutional neural networks are extracted and identified according to the video image frame of acquisition system
System carries out the shelter bridge characteristic area of video image frame with background characteristics area transmissions into shelter bridge working condition detection system
Comparison processing.
It handles preferably, the shelter bridge working condition judgment module is compared using frame differential method and obtains shelter bridge
Location status.
Preferably, the video image frame F that the convolutional neural networks system newly inputs current time tt, pass through first
Frame differential method obtains the black white image F after binaryzationFrame is poor, by black white image FFrame is poorBackground characteristics region be unified for black, will
Black white image FFrame is poorAircraft signature region be unified for white.
A kind of shelter bridge working state detecting method based on image processing techniques, method and step are as follows:
A, it obtains shelter bridge position sample database: the sample in airport in shelter bridge working region is acquired by video acquisition system
This video data, Sample video data include shelter bridge start by bridge Sample video data, shelter bridge by bridge complete Sample video data,
Shelter bridge, which initially moves off Sample video data and shelter bridge work, terminates Sample video data, and video acquisition system is by Sample video data
It is transmitted to data collection system, the data image frame acquisition module of data collection system carries out sample one by one to Sample video data
Video image frame acquires and sample video image frame is transmitted in data image frame electric wire division module, data image frame electricity
Sub-line division module delineates out electric wire A and electric wire B, the electric wire A for sample video image to sample video image frame
Frame is divided into the area A1 and the area A2, and the A2 zoning of sample video image frame is divided into the area A2B1 and the area A2B2, A2B1 by the electric wire B
Area is close to the area A1；Sample video image frame is extracted and identifies to obtain shelter bridge characteristic area and back by convolutional neural networks system
Scape characteristic area, and sample video image frame progress binary conversion treatment, feature are stored in shelter bridge position sample data after reunification
In library；
B, the real time video data in airport in shelter bridge working region, video acquisition system are acquired by video acquisition system
By real time video data and it is transmitted to data collection system, the data image frame acquisition module of data collection system is to real-time video
Data carry out real time video image frame acquisition one by one and real time video image frame are transmitted to data image frame electric wire division mould
In block, data image frame electric wire division module delineates out electric wire A and electric wire B, the electronics to real time video image frame
Real time video image frame is divided into the area A1 and the area A2 by line A, and the A2 zoning of real time video image frame is divided by the electric wire B
The area A2B1 and the area A2B2, the area A2B1 is close to the area A1；Real time video image frame is extracted and is identified by convolutional neural networks system
Obtain shelter bridge characteristic area and background characteristics region, and by real time video image frame carry out binary conversion treatment, feature after reunification with
The sample video image frame being stored in shelter bridge position sample database carries out image characteristics extraction and comparison processing；
C, shelter bridge operating state data is obtained, by shelter bridge working condition judgment module according to shelter bridge position sample database
Sample data the working condition of shelter bridge, the working condition of shelter bridge are obtained by image characteristics extraction, comparison processing, logic judgment
Including shelter bridge start by bridge, shelter bridge by bridge complete, shelter bridge initially move off with shelter bridge work terminate；Shelter bridge working condition judgment module
Shelter bridge working condition judgment method it is as follows:
If the shelter bridge characteristic area center line continuous N second of C1, real time video image frame is in the area A1, determine shelter bridge not
Work；
If the shelter bridge characteristic area center line of C2, real time video image frame enters the area A2B1 from the area A1, and continuously stops within N seconds
The area A2B1 is stayed in, then determines that the working condition of shelter bridge starts for shelter bridge by bridge；
If the shelter bridge characteristic area center line of C3, real time video image frame enters the area A2B2 from the area A2B1, and N seconds continuous
The area A2B2 is rested on, then determines that the working condition of shelter bridge is completed for shelter bridge by bridge；
If the shelter bridge characteristic area center line of C4, real time video image frame enters the area A2B1 from the area A2B2, and N seconds continuous
The area A2B1 is rested on, then determines that the working condition of shelter bridge initially moves off for shelter bridge；
If the shelter bridge characteristic area center line of C5, real time video image frame enters the area A1 from the area A2B1, and continuously stops within N seconds
The area A1 is stayed in, then determines that the working condition of shelter bridge terminates for shelter bridge work.
A kind of shelter bridge working state detecting method based on image processing techniques, method and step are as follows:
A, it obtains shelter bridge position sample database: the sample in airport in shelter bridge working region is acquired by video acquisition system
This video data, Sample video data include shelter bridge start by bridge Sample video data, shelter bridge by bridge complete Sample video data,
Shelter bridge, which initially moves off Sample video data and shelter bridge work, terminates Sample video data, and video acquisition system is by Sample video data
It is transmitted to data collection system, the data image frame acquisition module of data collection system carries out sample one by one to Sample video data
Video image frame acquires and sample video image frame is transmitted in data image frame electric wire division module, data image frame electricity
Sub-line division module delineates out electric wire A and electric wire B, the electric wire A for sample video image to sample video image frame
Frame is divided into the area A1 and the area A2, and the A2 zoning of sample video image frame is divided into the area A2B1 and the area A2B2, A2B1 by the electric wire B
Area is close to the area A1；Sample video image frame is extracted and identifies to obtain shelter bridge characteristic area and back by convolutional neural networks system
Scape characteristic area, and sample video image frame progress binary conversion treatment, feature are transmitted to the judgement of shelter bridge working condition after reunification
Module, shelter bridge working condition judgment module obtain the working condition of shelter bridge by image characteristics extraction, comparison processing, logic judgment
And it transmits and is stored in the sample database of aircraft position；At shelter bridge working condition judgment module judgement sample video image frame
Shelter bridge working condition judgment method it is as follows:
If the shelter bridge characteristic area center line continuous N second of A1, sample video image frame is in the area A1, determine shelter bridge not
Work；
If the shelter bridge characteristic area center line of A2, sample video image frame enters the area A2B1 from the area A1, and continuously stops within N seconds
The area A2B1 is stayed in, then determines that the working condition of shelter bridge starts for shelter bridge by bridge, it is then that the sample video image frame at this is corresponding
The shelter bridge for being stored as shelter bridge position sample database starts by bridge Sample video data；
If the shelter bridge characteristic area center line of A3, sample video image frame enters the area A2B2 from the area A2B1, and N seconds continuous
The area A2B2 is rested on, then determines that the working condition of shelter bridge is completed for shelter bridge by bridge, then by the sample video image frame pair at this
The shelter bridge that should be stored as shelter bridge position sample database completes Sample video data by bridge；
If the shelter bridge characteristic area center line of A4, sample video image frame enters the area A2B1 from the area A2B2, and N seconds continuous
The area A2B1 is rested on, then determines that the working condition of shelter bridge initially moves off for shelter bridge, then by the sample video image frame pair at this
The shelter bridge that should be stored as shelter bridge position sample database initially moves off Sample video data；
If the shelter bridge characteristic area center line of A5, sample video image frame enters the area A1 from the area A2B1, and continuously stops within N seconds
The area A1 is stayed in, then determines that the working condition of shelter bridge terminates for shelter bridge work, then deposits the sample video image frame correspondence at this
Storage is that the shelter bridge work of shelter bridge position sample database terminates Sample video data；
B, the real time video data in airport in shelter bridge working region, video acquisition system are acquired by video acquisition system
By real time video data and it is transmitted to data collection system, the data image frame acquisition module of data collection system is to real-time video
Data carry out real time video image frame acquisition one by one and real time video image frame are transmitted to data image frame electric wire division mould
In block, data image frame electric wire division module delineates out electric wire A and electric wire B, the electronics to real time video image frame
Real time video image frame is divided into the area A1 and the area A2 by line A, and the A2 zoning of real time video image frame is divided by the electric wire B
The area A2B1 and the area A2B2, the area A2B1 is close to the area A1；Real time video image frame is extracted and is identified by convolutional neural networks system
Obtain shelter bridge characteristic area and background characteristics region, and by real time video image frame carry out binary conversion treatment, image characteristics extraction,
Feature carries out aspect ratio to processing with the sample video image frame being stored in shelter bridge position sample database after reunification；
C, shelter bridge operating state data is obtained, by shelter bridge working condition judgment module according to shelter bridge position sample database
Sample data the working condition of shelter bridge, the working condition of shelter bridge are obtained by image characteristics extraction, comparison processing, logic judgment
Including shelter bridge start by bridge, shelter bridge by bridge complete, shelter bridge initially move off with shelter bridge work terminate；Shelter bridge working condition judgment module
Shelter bridge working condition judgment method it is as follows:
If the shelter bridge characteristic area center line continuous N second of C1, real time video image frame is in the area A1, determine shelter bridge not
Work；
If the shelter bridge characteristic area center line of C2, real time video image frame enters the area A2B1 from the area A1, and continuously stops within N seconds
The area A2B1 is stayed in, then determines that the working condition of shelter bridge starts for shelter bridge by bridge；
If the shelter bridge characteristic area center line of C3, real time video image frame enters the area A2B2 from the area A2B1, and N seconds continuous
The area A2B2 is rested on, then determines that the working condition of shelter bridge is completed for shelter bridge by bridge；
If the shelter bridge characteristic area center line of C4, real time video image frame enters the area A2B1 from the area A2B2, and N seconds continuous
The area A2B1 is rested on, then determines that the working condition of shelter bridge initially moves off for shelter bridge；
If the shelter bridge characteristic area center line of C5, real time video image frame enters the area A1 from the area A2B1, and continuously stops within N seconds
The area A1 is stayed in, then determines that the working condition of shelter bridge terminates for shelter bridge work.
Shelter bridge start by bridge, shelter bridge by bridge complete, shelter bridge initially move off with shelter bridge work terminate
Video acquisition system, data collection system, shelter bridge position detection model framework, data image frame electric wire divide mould
Block, data image frame acquisition module, convolutional neural networks system, shelter bridge position sample database, shelter bridge working condition detection system
System, shelter bridge working condition judgment module.
Preferably, the integer that M is 10 or more in the step C, N is the integer between 5 to 10 in the step C.
Preferably, in the case of Sample video data include night, sleet, mist or haze, dust and sand weather in the step A
Shelter bridge position sample data, wherein the shelter bridge position sample data accounting in the case of night weather is no less than 10%, moderate rain
Shelter bridge position sample data accounting in the case of snowy day gas is no less than 10%, wherein the shelter bridge position in the case of mist or haze weather
Sample data accounting is no less than 10%, and wherein the shelter bridge position sample data accounting in the case of dust and sand weather is no less than 10%.
Preferably, the video acquisition system of the step A and step B includes several video cameras, all videos
Camera be respectively arranged in the front of shelter bridge working region in airport, left front, right front, surface, upper left side and it is rear on
Set place in orientation；The data collection system of the step A and step B acquires the single video data of all video cameras and combines
At video data transmission into data image frame electric wire division module；The video camera is digital camera, and digital
The resolution ratio of video camera is not less than 1080 × 720.
Preferably, the picture number of the sample video image frame of the shelter bridge position sample database is no less than 10,000
?.
The present invention compared with the prior art, have the following advantages that and the utility model has the advantages that
The present invention judges the current working condition of shelter bridge by automatically analyzing to machine level ground monitor video signal, can be simultaneously
The beginning and deadline that beginning and deadline and shelter bridge of the shelter bridge by machine are disembarked are obtained, so that controlling officer is to aircraft
Lasting scene guidance is carried out, provides correct time foundation for air station flight guarantee.
Detailed description of the invention
Fig. 1 is the principle of the present invention structural block diagram.
Specific embodiment
The present invention is described in further detail below with reference to embodiment:
Embodiment one
As shown in Figure 1, a kind of shelter bridge working condition detection system based on image processing techniques, including video acquisition system
System, data collection system, convolutional neural networks system and shelter bridge working condition detection system, the video acquisition system, data
Acquisition system, convolutional neural networks system, shelter bridge working condition detection system successively communicate to connect, the data collection system packet
Data image frame electric wire division module and data image frame acquisitions module are included, the convolutional neural networks internal system has corridor
Bridge location sets sample database, and the shelter bridge working condition detection system includes shelter bridge working condition judgment module；The video is adopted
Collecting system includes several video cameras, and the video acquisition system is used to acquire video data and combination of transmitted to data are adopted
Collecting system, the data image frame acquisition module of the data collection system are used to carry out video data video image frame one by one to adopt
Collect and video image frame is transmitted to data image frame electric wire division module, the data image frame electric wire division module is used
In delineating out electric wire A to video image frame and video image frame is divided into the area A1 and the area A2 by electric wire B, the electric wire A,
The A2 zoning of video image frame is divided into the area A2B1 and the area A2B2 by the electric wire B, and the area A2B1 is close to the area A1；The shelter bridge position
Sample database is used to be stored with the sample data of shelter bridge working condition, and the sample data includes that shelter bridge starts by bridge sample number
According to, shelter bridge by bridge completed sample according to, shelter bridge initially moves off sample data and shelter bridge work terminates sample data；The convolution mind
Through network system be used for data collection system collected video image frame and shelter bridge position sample database sample number
According to image characteristics extraction and comparison processing is carried out, the shelter bridge working condition judgment module is for according to shelter bridge position sample data
The sample data in library obtains the working condition of shelter bridge, the work shape of shelter bridge by image characteristics extraction, comparison processing, logic judgment
State include shelter bridge start by bridge, shelter bridge by bridge complete, shelter bridge initially move off with shelter bridge work terminate.
Convolutional neural networks system is for extracting the video image frame of data collection system and identifying shelter bridge characteristic area
Domain and background characteristics region, the convolutional neural networks system is by the shelter bridge characteristic area of video image frame and background characteristics region
It is transmitted in shelter bridge working condition detection system and compares processing.
Shelter bridge working condition judgment module is compared using frame differential method to be handled and obtains shelter bridge location status.
The video image frame F that convolutional neural networks system newly inputs current time tt, obtained first by frame differential method
Black white image F after to binaryzationFrame is poor, by black white image FFrame is poorBackground characteristics region be unified for black, by black white image FFrame is poor's
Aircraft signature region is unified for white.
A kind of shelter bridge working state detecting method based on image processing techniques, method and step are as follows:
A, it obtains shelter bridge position sample database: the sample in airport in shelter bridge working region is acquired by video acquisition system
This video data, Sample video data include shelter bridge start by bridge Sample video data, shelter bridge by bridge complete Sample video data,
Shelter bridge, which initially moves off Sample video data and shelter bridge work, terminates Sample video data, and video acquisition system is by Sample video data
It is transmitted to data collection system, the data image frame acquisition module of data collection system carries out sample one by one to Sample video data
Video image frame acquires and sample video image frame is transmitted in data image frame electric wire division module, data image frame electricity
Sub-line division module delineates out electric wire A and electric wire B, the electric wire A for sample video image to sample video image frame
Frame is divided into the area A1 and the area A2, and the A2 zoning of sample video image frame is divided into the area A2B1 and the area A2B2, A2B1 by the electric wire B
Area is close to the area A1；Sample video image frame is extracted and identifies to obtain shelter bridge characteristic area and back by convolutional neural networks system
Scape characteristic area, and sample video image frame progress binary conversion treatment, feature are stored in shelter bridge position sample data after reunification
In library；The shelter bridge of the present embodiment shelter bridge position sample database, which starts to be directed to by bridge Sample video data, to be had determined as
(the Sample video data are when having determined as shelter bridge to start by bridge to Sample video data when shelter bridge starts by bridge, and video is adopted
The collected Sample video data of collecting system institute)；The shelter bridge of the present embodiment shelter bridge position sample database completes sample view by bridge
According to the Sample video data for having determined as when shelter bridge is completed by bridge are directed to, (the Sample video data are really to frequency
When being set to shelter bridge by bridge completion, the collected Sample video data of video acquisition system institute)；The present embodiment shelter bridge position sample number
Sample video data, which are initially moved off, according to the shelter bridge in library is directed to the Sample video number having determined as when shelter bridge initially moves off
According to (the Sample video data are video acquisition system institute collected Sample video number when having determined as shelter bridge to initially move off
According to)；The shelter bridge work end Sample video data of the present embodiment shelter bridge position sample database, which are directed to, to be had determined as
(the Sample video data are at the end of having determined as shelter bridge work to Sample video data at the end of shelter bridge work, and video is adopted
The collected Sample video data of collecting system institute).Currently preferred Sample video data include night, sleet, mist or haze,
Shelter bridge position sample data in the case of dust and sand weather, wherein the shelter bridge position sample data accounting in the case of night weather is many
In 10%, wherein the shelter bridge position sample data accounting in the case of rain and snow weather is no less than 10%, wherein mist or haze weather feelings
Shelter bridge position sample data accounting under condition is no less than 10%, wherein the shelter bridge position sample data accounting in the case of dust and sand weather
No less than 10%.The picture number of the sample video image frame of shelter bridge position sample database of the present invention is no less than 10,000.
B, the real time video data in airport in shelter bridge working region, video acquisition system are acquired by video acquisition system
By real time video data and it is transmitted to data collection system, the data image frame acquisition module of data collection system is to real-time video
Data carry out real time video image frame acquisition one by one and real time video image frame are transmitted to data image frame electric wire division mould
In block, data image frame electric wire division module delineates out electric wire A and electric wire B, the electronics to real time video image frame
Real time video image frame is divided into the area A1 and the area A2 by line A, and the A2 zoning of real time video image frame is divided by the electric wire B
The area A2B1 and the area A2B2, the area A2B1 is close to the area A1；Real time video image frame is extracted and is identified by convolutional neural networks system
Obtain shelter bridge characteristic area and background characteristics region, and by real time video image frame carry out binary conversion treatment, feature after reunification with
The sample video image frame being stored in shelter bridge position sample database carries out image characteristics extraction and comparison processing.
C, shelter bridge operating state data is obtained, by shelter bridge working condition judgment module according to shelter bridge position sample database
Sample data the working condition of shelter bridge, the working condition of shelter bridge are obtained by image characteristics extraction, comparison processing, logic judgment
Including shelter bridge start by bridge, shelter bridge by bridge complete, shelter bridge initially move off with shelter bridge work terminate；Shelter bridge working condition judgment module
Shelter bridge working condition judgment method it is as follows:
If the shelter bridge characteristic area center line continuous N second of C1, real time video image frame is in the area A1, determine shelter bridge not
Work；
If the shelter bridge characteristic area center line of C2, real time video image frame enters the area A2B1 from the area A1, and continuously stops within N seconds
The area A2B1 is stayed in, then determines that the working condition of shelter bridge starts for shelter bridge by bridge；
If the shelter bridge characteristic area center line of C3, real time video image frame enters the area A2B2 from the area A2B1, and N seconds continuous
The area A2B2 is rested on, then determines that the working condition of shelter bridge is completed for shelter bridge by bridge；
If the shelter bridge characteristic area center line of C4, real time video image frame enters the area A2B1 from the area A2B2, and N seconds continuous
The area A2B1 is rested on, then determines that the working condition of shelter bridge initially moves off for shelter bridge；
If the shelter bridge characteristic area center line of C5, real time video image frame enters the area A1 from the area A2B1, and continuously stops within N seconds
The area A1 is stayed in, then determines that the working condition of shelter bridge terminates for shelter bridge work.
The integer that M is 10 or more in step C of the present invention, N is the integer between 5 to 10 in the step C.
The video acquisition system of step A and step B of the present invention include several video cameras, all video cameras point
It is not installed at the front of shelter bridge working region in airport, left front, right front, surface, upper left side and back upper place position；
The data collection system of the step A and step B acquires the single video data of all video cameras and is combined into video counts
According to being transmitted in data image frame electric wire division module；The video camera is digital camera, and digital camera
Resolution ratio is not less than 1080 × 720.
Embodiment two
A kind of shelter bridge working state detecting method based on image processing techniques, method and step are as follows:
A, it obtains shelter bridge position sample database: the sample in airport in shelter bridge working region is acquired by video acquisition system
This video data, Sample video data include shelter bridge start by bridge Sample video data, shelter bridge by bridge complete Sample video data,
Shelter bridge, which initially moves off Sample video data and shelter bridge work, terminates Sample video data, and video acquisition system is by Sample video data
It is transmitted to data collection system, the data image frame acquisition module of data collection system carries out sample one by one to Sample video data
Video image frame acquires and sample video image frame is transmitted in data image frame electric wire division module, data image frame electricity
Sub-line division module delineates out electric wire A and electric wire B, the electric wire A for sample video image to sample video image frame
Frame is divided into the area A1 and the area A2, and the A2 zoning of sample video image frame is divided into the area A2B1 and the area A2B2, A2B1 by the electric wire B
Area is close to the area A1；Sample video image frame is extracted and identifies to obtain shelter bridge characteristic area and back by convolutional neural networks system
Scape characteristic area, and sample video image frame progress binary conversion treatment, feature are transmitted to the judgement of shelter bridge working condition after reunification
Module, shelter bridge working condition judgment module obtain the working condition of shelter bridge by image characteristics extraction, comparison processing, logic judgment
And it transmits and is stored in the sample database of aircraft position；The shelter bridge of the present embodiment shelter bridge position sample database starts by bridge sample
Video data can be determined as Sample video data when shelter bridge starts by bridge from shelter bridge working condition judgment module, certainly
Sample video data when can also start from shelter bridge is had determined as by bridge；The present embodiment shelter bridge position sample database
Shelter bridge by bridge complete Sample video data can from shelter bridge working condition judgment module be determined as shelter bridge by bridge complete when
Sample video data, naturally it is also possible to from have determined as shelter bridge by bridge complete when Sample video data；This implementation
The shelter bridge of example shelter bridge position sample database, which initially moves off Sample video data, can derive from shelter bridge working condition judgment module
It is determined as Sample video data when shelter bridge initially moves off, naturally it is also possible to from having determined as when shelter bridge initially moves off
Sample video data；The shelter bridge work of the present embodiment shelter bridge position sample database, which terminates Sample video data, can derive from corridor
Bridge working condition judgment module is determined as the Sample video data at the end of shelter bridge work, naturally it is also possible to from having determined
For the Sample video data at the end of shelter bridge work.At shelter bridge working condition judgment module judgement sample video image frame
The judgment method of shelter bridge working condition is as follows:
If the shelter bridge characteristic area center line continuous N second of A1, sample video image frame is in the area A1, determine shelter bridge not
Work；
If the shelter bridge characteristic area center line of A2, sample video image frame enters the area A2B1 from the area A1, and continuously stops within N seconds
The area A2B1 is stayed in, then determines that the working condition of shelter bridge starts for shelter bridge by bridge, it is then that the sample video image frame at this is corresponding
The shelter bridge for being stored as shelter bridge position sample database starts by bridge Sample video data；
If the shelter bridge characteristic area center line of A3, sample video image frame enters the area A2B2 from the area A2B1, and N seconds continuous
The area A2B2 is rested on, then determines that the working condition of shelter bridge is completed for shelter bridge by bridge, then by the sample video image frame pair at this
The shelter bridge that should be stored as shelter bridge position sample database completes Sample video data by bridge；
If the shelter bridge characteristic area center line of A4, sample video image frame enters the area A2B1 from the area A2B2, and N seconds continuous
The area A2B1 is rested on, then determines that the working condition of shelter bridge initially moves off for shelter bridge, then by the sample video image frame pair at this
The shelter bridge that should be stored as shelter bridge position sample database initially moves off Sample video data；
If the shelter bridge characteristic area center line of A5, sample video image frame enters the area A1 from the area A2B1, and continuously stops within N seconds
The area A1 is stayed in, then determines that the working condition of shelter bridge terminates for shelter bridge work, then deposits the sample video image frame correspondence at this
Storage is that the shelter bridge work of shelter bridge position sample database terminates Sample video data；
Sample video data of the invention include the shelter bridge position sample in the case of night, sleet, mist or haze, dust and sand weather
Notebook data, wherein the shelter bridge position sample data accounting in the case of night weather is no less than 10%, wherein in the case of rain and snow weather
Shelter bridge position sample data accounting be no less than 10%, the wherein shelter bridge position sample data accounting in the case of mist or haze weather
No less than 10%, wherein the shelter bridge position sample data accounting in the case of dust and sand weather is no less than 10%.Shelter bridge position of the present invention
The picture number of the sample video image frame of sample database is no less than 10,000.
B, the real time video data in airport in shelter bridge working region, video acquisition system are acquired by video acquisition system
By real time video data and it is transmitted to data collection system, the data image frame acquisition module of data collection system is to real-time video
Data carry out real time video image frame acquisition one by one and real time video image frame are transmitted to data image frame electric wire division mould
In block, data image frame electric wire division module delineates out electric wire A and electric wire B, the electronics to real time video image frame
Real time video image frame is divided into the area A1 and the area A2 by line A, and the A2 zoning of real time video image frame is divided by the electric wire B
The area A2B1 and the area A2B2, the area A2B1 is close to the area A1；Real time video image frame is extracted and is identified by convolutional neural networks system
Obtain shelter bridge characteristic area and background characteristics region, and by real time video image frame carry out binary conversion treatment, image characteristics extraction,
Feature carries out aspect ratio to processing, convolution mind with the sample video image frame being stored in shelter bridge position sample database after reunification
Real time video image frame and sample video image frame are compared through network system and obtain the primary work state conclusion of shelter bridge.
C, shelter bridge operating state data is obtained, by shelter bridge working condition judgment module according to shelter bridge position sample database
Sample data the working condition of shelter bridge, the working condition of shelter bridge are obtained by image characteristics extraction, comparison processing, logic judgment
Including shelter bridge start by bridge, shelter bridge by bridge complete, shelter bridge initially move off with shelter bridge work terminate；Shelter bridge working condition judgment module
Shelter bridge working condition judgment method it is as follows:
If the shelter bridge characteristic area center line continuous N second of C1, real time video image frame is in the area A1, determine shelter bridge not
Work；
If the shelter bridge characteristic area center line of C2, real time video image frame enters the area A2B1 from the area A1, and continuously stops within N seconds
The area A2B1 is stayed in, then determines that the working condition of shelter bridge starts for shelter bridge by bridge；
If the shelter bridge characteristic area center line of C3, real time video image frame enters the area A2B2 from the area A2B1, and N seconds continuous
The area A2B2 is rested on, then determines that the working condition of shelter bridge is completed for shelter bridge by bridge；
If the shelter bridge characteristic area center line of C4, real time video image frame enters the area A2B1 from the area A2B2, and N seconds continuous
The area A2B1 is rested on, then determines that the working condition of shelter bridge initially moves off for shelter bridge；
If the shelter bridge characteristic area center line of C5, real time video image frame enters the area A1 from the area A2B1, and continuously stops within N seconds
The area A1 is stayed in, then determines that the working condition of shelter bridge terminates for shelter bridge work.
The integer that M is 10 or more in step C, N is the integer between 5 to 10 in the step C.
Shelter bridge working condition detection system of the invention for shelter bridge working condition above-mentioned logic judgment and combine preliminary
Working condition conclusion simultaneously finally obtains shelter bridge operating state data.
The video acquisition system of step A and step B of the present invention include several video cameras, all video cameras point
It is not installed at the front of shelter bridge working region in airport, left front, right front, surface, upper left side and back upper place position；
The data collection system of the step A and step B acquires the single video data of all video cameras and is combined into video counts
According to being transmitted in data image frame electric wire division module；The video camera is digital camera, and digital camera
Resolution ratio is not less than 1080 × 720.
Embodiment three
A kind of shelter bridge working state detecting method based on image processing techniques, method and step are as follows:
A, the shelter bridge picture sample library for being used for neural metwork training is established.
B, convolutional neural networks of the training for shelter bridge identification, hereinafter referred to as identification network.
C, shelter bridge detection model frame, hereinafter referred to as detection framework are constructed.
D, the detection framework is connected with the data acquisition module of a responsible collection monitoring video.
E, video data is transferred to detection framework frame by frame by data acquisition module.
F, the detection framework handles each frame video image of input, ultimately generates shelter bridge testing result, i.e.,
Shelter bridge detection block.
According to the position of shelter bridge detection block, judge that shelter bridge working condition, the working condition of shelter bridge include not working, starting to lean on
Bridge (starting close to aircraft) is completed by bridge, starts to remove bridge (initially moving off aircraft), removes bridge completion.
Optionally, the shelter bridge working condition deterministic process comprises the steps of:
Electric wire A is divided in monitor video, machine level ground monitoring image is divided into two regions A1 and A2 by A.At shelter bridge
When non-job state, the central point of shelter bridge detection block should be always positioned at the area A1.
Electric wire B is divided in monitor video, A2 zoning is divided into the area A2B1 and the area A2B2 by B.
If the testing result frame central point of shelter bridge enters the area A2B1 from the area A1, and continuously rests on the area A2 within N seconds, then by shelter bridge
Working condition is set as starting by bridge.
If the shelter bridge of working condition enters the area A2B2 from the area A2B1, and continuously rests on the area A2B2 within N seconds, then it is assumed that shelter bridge leans on
Bridge is completed.
If entering the area A2B1 from the area A2B2 by the shelter bridge of machine state, and continuously rest on the area A2B1 within N seconds, then it is assumed that shelter bridge from
Movement of flying an airplanepilot an airplane is completed.
If shelter bridge enters the area A1 from the area A2B1, and continuously rests on the area A1 within N seconds, then it is assumed that shelter bridge work terminates.
The value of the present embodiment N is preferably the integer between 5 to 10.
It records and externally exports shelter bridge working condition.
Shelter bridge picture sample library establishment process comprises the steps of:
From the monitor video of airport machine level ground, a certain number of pictures comprising shelter bridge are acquired, and be labeled to picture.It can
Choosing, the picture number in sample database is no less than 10,000.Optionally, Ying Xuanyong digital camera color image collected,
And the resolution ratio of digital camera is not less than 1080*720.Optionally, to guarantee system can be stable under different perspectives pair
Shelter bridge identified, should be respectively from the view for being installed on machine level ground front, left front, right front, surface, upper left side, back upper place
Picture is acquired in frequency monitoring device, and distinct device picture number collected is identical.It preferably, is guarantee neural network to night
Between, the adaptability of sleet, mist or haze, dust and sand weather, a part in selected picture should be under above-mentioned weather condition
Acquisition.Preferably, the picture collected under night, rainy day, snowy day, greasy weather or haze sky scene, should account for picture sample respectively
10% or more of library.
The object detection convolutional neural networks training process of the present embodiment comprises the steps of:
Using the training sample marked, object detection convolutional neural networks are trained.Preferably, the sample mark
Injecting method and neural network training method can be according to Redmon J, Divvala S K, Girshick R B, et al.You
Only Look Once:Unified,Real-Time Object Detection[J].computer vision and
The training of pattern recognition, 2016:779-788. literature method.
The frame differential method of the present embodiment comprises the steps of:
By mixed Gaussian background modeling method, background model is modeled and is updated.
The video image F that current time t is newly inputtedt, first by frame difference method, the black white image after obtaining binaryzation
FFrame is poor, wherein background area is black, and foreground area is white.Background modeling, update and calculating frame difference image FFrame is poorMethod,
Can according to Harville M.A Framework for High-Level Feedback to Adaptive, Per-Pixel,
Mixture-of-Gaussian Background Models [C], European Conference on Computer
Vision.2002. the frame difference image method in document.
From frame difference image FFrame is poorIn, foreground area is extracted, and using each region as a couple candidate detection region.It is poor from frame
The method that foreground area is extracted in image afterwards, method is according to Xu W, Zhao Q, Gu D.Fragmentation
handling for visual tracking[J].Signal Image&Video Processing,2014,8(8):1639-
Method in 1649. documents extracts foreground area.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Made any modifications, equivalent replacements, and improvements etc., should all be included in the protection scope of the present invention within mind and principle.
Claims (10)
1. a kind of shelter bridge working condition detection system based on image processing techniques, it is characterised in that: including video acquisition system,
Data collection system, convolutional neural networks system and shelter bridge working condition detection system, the video acquisition system, data acquisition
System, convolutional neural networks system, shelter bridge working condition detection system successively communicate to connect, and the data collection system includes number
According to picture frame electric wire division module and data image frame acquisitions module, the convolutional neural networks internal system has shelter bridge position
Sample database is set, the shelter bridge working condition detection system includes shelter bridge working condition judgment module；The video acquisition system
System includes several video cameras, and for acquiring video data, simultaneously combination of transmitted to data acquisition is the video acquisition system
System, the data image frame acquisition module of the data collection system are used to carry out video data video image frame acquisition one by one simultaneously
Video image frame is transmitted to data image frame electric wire division module, the data image frame electric wire division module for pair
Video image frame delineates out electric wire A and electric wire B, and video image frame is divided into the area A1 and the area A2 by the electric wire A, described
The A2 zoning of video image frame is divided into the area A2B1 and the area A2B2 by electric wire B, and the area A2B1 is close to the area A1；The shelter bridge position sample
Database is used to be stored with the sample data of shelter bridge working condition, the sample data include shelter bridge start by bridge sample data,
Shelter bridge, which initially moves off sample data and shelter bridge work by bridge completed sample evidence, shelter bridge, terminates sample data；The convolutional Neural
Network system be used for data collection system collected video image frame and shelter bridge position sample database sample data
It carries out image characteristics extraction and comparison processing, the shelter bridge working condition judgment module is used for according to shelter bridge position sample database
Sample data the working condition of shelter bridge, the working condition of shelter bridge are obtained by image characteristics extraction, comparison processing, logic judgment
Including shelter bridge start by bridge, shelter bridge by bridge complete, shelter bridge initially move off with shelter bridge work terminate.
2. a kind of shelter bridge working condition detection system based on image processing techniques described in accordance with the claim 1, feature exist
In: the convolutional neural networks system is for extracting the video image frame of data collection system and identifying shelter bridge characteristic area
With background characteristics region, the convolutional neural networks system passes the shelter bridge characteristic area of video image frame and background characteristics region
It transports in shelter bridge working condition detection system and compares processing.
3. a kind of shelter bridge working condition detection system based on image processing techniques described in accordance with the claim 1, feature exist
In: the shelter bridge working condition judgment module compares processing using frame differential method and obtains shelter bridge location status.
4. a kind of shelter bridge working condition detection system based on image processing techniques described in accordance with the claim 1, feature exist
In: the video image frame F that the convolutional neural networks system newly inputs current time tt, obtained first by frame differential method
Black white image F after binaryzationFrame is poor, by black white image FFrame is poorBackground characteristics region be unified for black, by black white image FFrame is poorFly
Machine characteristic area is unified for white.
5. a kind of shelter bridge working state detecting method based on image processing techniques, it is characterised in that: its method and step is as follows:
A, it obtains shelter bridge position sample database: acquiring the sample in airport in shelter bridge working region by video acquisition system and regard
Frequency evidence, Sample video data include that shelter bridge starts to complete Sample video data, shelter bridge by bridge by bridge Sample video data, shelter bridge
Initially moving off Sample video data and shelter bridge work terminates Sample video data, and video acquisition system transmits Sample video data
To data collection system, the data image frame acquisition module of data collection system carries out Sample video one by one to Sample video data
Sample video image frame is simultaneously transmitted in data image frame electric wire division module, data image frame electric wire by image frame acquisitions
Division module delineates out electric wire A and electric wire B, the electric wire A to sample video image frame and draws sample video image frame
It is divided into the area A1 and the area A2, the A2 zoning of sample video image frame is divided into the area A2B1 and the area A2B2 by the electric wire B, and the area A2B1 is leaned on
The nearly area A1；Sample video image frame is extracted and identifies to obtain shelter bridge characteristic area and background spy by convolutional neural networks system
Region is levied, and sample video image frame progress binary conversion treatment, feature are stored in after reunification in shelter bridge position sample database；
B, the real time video data in airport in shelter bridge working region is acquired by video acquisition system, video acquisition system will be real
When video data and be transmitted to data collection system, the data image frame acquisition module of data collection system is to real time video data
Real time video image frame one by one is carried out to acquire and real time video image frame is transmitted in data image frame electric wire division module,
Data image frame electric wire division module delineates out electric wire A to real time video image frame and electric wire B, the electric wire A will
Real time video image frame is divided into the area A1 and the area A2, and the A2 zoning of real time video image frame is divided into the area A2B1 by the electric wire B
With the area A2B2, the area A2B1 is close to the area A1；Real time video image frame is extracted and identifies to obtain corridor by convolutional neural networks system
Bridge characteristic area and background characteristics region, and by real time video image frame carry out binary conversion treatment, feature after reunification be stored in
Sample video image frame in shelter bridge position sample database carries out image characteristics extraction and comparison processing；
C, shelter bridge operating state data is obtained, by shelter bridge working condition judgment module according to the sample of shelter bridge position sample database
Notebook data show that the working condition of shelter bridge, the working condition of shelter bridge include by image characteristics extraction, comparison processing, logic judgment
Shelter bridge start by bridge, shelter bridge by bridge complete, shelter bridge initially move off with shelter bridge work terminate；The corridor of shelter bridge working condition judgment module
Bridge working condition judgment method is as follows:
If the shelter bridge characteristic area center line continuous N second of C1, real time video image frame is in the area A1, the non-work of shelter bridge is determined
Make；
If the shelter bridge characteristic area center line of C2, real time video image frame enters the area A2B1 from the area A1, and continuously rests within N seconds
The area A2B1 then determines that the working condition of shelter bridge starts for shelter bridge by bridge；
If the shelter bridge characteristic area center line of C3, real time video image frame enters the area A2B2 from the area A2B1, and continuous N seconds stop
In the area A2B2, then determine that the working condition of shelter bridge is completed for shelter bridge by bridge；
If the shelter bridge characteristic area center line of C4, real time video image frame enters the area A2B1 from the area A2B2, and continuous N seconds stop
In the area A2B1, then determine that the working condition of shelter bridge initially moves off for shelter bridge；
If the shelter bridge characteristic area center line of C5, real time video image frame enters the area A1 from the area A2B1, and continuously rests within N seconds
The area A1 then determines that the working condition of shelter bridge terminates for shelter bridge work.
6. a kind of shelter bridge working state detecting method based on image processing techniques, it is characterised in that: its method and step is as follows:
A, it obtains shelter bridge position sample database: acquiring the sample in airport in shelter bridge working region by video acquisition system and regard
Frequency evidence, Sample video data include that shelter bridge starts to complete Sample video data, shelter bridge by bridge by bridge Sample video data, shelter bridge
Initially moving off Sample video data and shelter bridge work terminates Sample video data, and video acquisition system transmits Sample video data
To data collection system, the data image frame acquisition module of data collection system carries out Sample video one by one to Sample video data
Sample video image frame is simultaneously transmitted in data image frame electric wire division module, data image frame electric wire by image frame acquisitions
Division module delineates out electric wire A and electric wire B, the electric wire A to sample video image frame and draws sample video image frame
It is divided into the area A1 and the area A2, the A2 zoning of sample video image frame is divided into the area A2B1 and the area A2B2 by the electric wire B, and the area A2B1 is leaned on
The nearly area A1；Sample video image frame is extracted and identifies to obtain shelter bridge characteristic area and background spy by convolutional neural networks system
Region is levied, and sample video image frame progress binary conversion treatment, feature are transmitted to shelter bridge working condition judgment module after reunification,
Shelter bridge working condition judgment module obtains the working condition and biography of shelter bridge by image characteristics extraction, comparison processing, logic judgment
It is defeated to be stored in the sample database of aircraft position；Corridor at shelter bridge working condition judgment module judgement sample video image frame
The judgment method of bridge working condition is as follows:
If the shelter bridge characteristic area center line continuous N second of A1, sample video image frame is in the area A1, the non-work of shelter bridge is determined
Make；
If the shelter bridge characteristic area center line of A2, sample video image frame enters the area A2B1 from the area A1, and continuously rests within N seconds
The area A2B1 then determines that the working condition of shelter bridge starts for shelter bridge by bridge, then by the corresponding storage of sample video image frame at this
Start for the shelter bridge of shelter bridge position sample database by bridge Sample video data；
If the shelter bridge characteristic area center line of A3, sample video image frame enters the area A2B2 from the area A2B1, and continuous N seconds stop
In the area A2B2, then determine that the working condition of shelter bridge is completed for shelter bridge by bridge, then deposits the sample video image frame correspondence at this
Storage is that the shelter bridge of shelter bridge position sample database completes Sample video data by bridge；
If the shelter bridge characteristic area center line of A4, sample video image frame enters the area A2B1 from the area A2B2, and continuous N seconds stop
In the area A2B1, then determine that the working condition of shelter bridge initially moves off for shelter bridge, then deposits the sample video image frame correspondence at this
Storage is that the shelter bridge of shelter bridge position sample database initially moves off Sample video data；
If the shelter bridge characteristic area center line of A5, sample video image frame enters the area A1 from the area A2B1, and continuously rests within N seconds
The area A1 then determines that the working condition of shelter bridge terminates for shelter bridge work, is then stored as the sample video image frame correspondence at this
The shelter bridge work of shelter bridge position sample database terminates Sample video data；
B, the real time video data in airport in shelter bridge working region is acquired by video acquisition system, video acquisition system will be real
When video data and be transmitted to data collection system, the data image frame acquisition module of data collection system is to real time video data
Real time video image frame one by one is carried out to acquire and real time video image frame is transmitted in data image frame electric wire division module,
Data image frame electric wire division module delineates out electric wire A to real time video image frame and electric wire B, the electric wire A will
Real time video image frame is divided into the area A1 and the area A2, and the A2 zoning of real time video image frame is divided into the area A2B1 by the electric wire B
With the area A2B2, the area A2B1 is close to the area A1；Real time video image frame is extracted and identifies to obtain corridor by convolutional neural networks system
Bridge characteristic area and background characteristics region, and real time video image frame is subjected to binary conversion treatment, image characteristics extraction, feature system
Aspect ratio is carried out to processing with the sample video image frame being stored in shelter bridge position sample database after one；
C, shelter bridge operating state data is obtained, by shelter bridge working condition judgment module according to the sample of shelter bridge position sample database
Notebook data show that the working condition of shelter bridge, the working condition of shelter bridge include by image characteristics extraction, comparison processing, logic judgment
Shelter bridge start by bridge, shelter bridge by bridge complete, shelter bridge initially move off with shelter bridge work terminate；The corridor of shelter bridge working condition judgment module
Bridge working condition judgment method is as follows:
If the shelter bridge characteristic area center line continuous N second of C1, real time video image frame is in the area A1, the non-work of shelter bridge is determined
Make；
If the shelter bridge characteristic area center line of C2, real time video image frame enters the area A2B1 from the area A1, and continuously rests within N seconds
The area A2B1 then determines that the working condition of shelter bridge starts for shelter bridge by bridge；
If the shelter bridge characteristic area center line of C3, real time video image frame enters the area A2B2 from the area A2B1, and continuous N seconds stop
In the area A2B2, then determine that the working condition of shelter bridge is completed for shelter bridge by bridge；
If the shelter bridge characteristic area center line of C4, real time video image frame enters the area A2B1 from the area A2B2, and continuous N seconds stop
In the area A2B1, then determine that the working condition of shelter bridge initially moves off for shelter bridge；
If the shelter bridge characteristic area center line of C5, real time video image frame enters the area A1 from the area A2B1, and continuously rests within N seconds
The area A1 then determines that the working condition of shelter bridge terminates for shelter bridge work.
7. according to a kind of shelter bridge working condition detection system based on image processing techniques described in claim 5 or 6, feature
Be: the integer that M is 10 or more in the step C, N is the integer between 5 to 10 in the step C.
8. according to a kind of shelter bridge working condition detection system based on image processing techniques described in claim 5 or 6, feature
Be: Sample video data include the shelter bridge position sample in the case of night, sleet, mist or haze, dust and sand weather in the step A
Notebook data, wherein the shelter bridge position sample data accounting in the case of night weather is no less than 10%, wherein in the case of rain and snow weather
Shelter bridge position sample data accounting be no less than 10%, the wherein shelter bridge position sample data accounting in the case of mist or haze weather
No less than 10%, wherein the shelter bridge position sample data accounting in the case of dust and sand weather is no less than 10%.
9. according to a kind of shelter bridge working condition detection system based on image processing techniques described in claim 5 or 6, feature
Be: the video acquisition system of the step A and step B includes several video cameras, and all video cameras are pacified respectively
Loaded in airport at the front of shelter bridge working region, left front, right front, surface, upper left side and back upper place position；It is described
The data collection system of step A and step B acquire the single video data of all video cameras and are combined into video data biography
It transports in data image frame electric wire division module；The video camera is digital camera, and the resolution of digital camera
Rate is not less than 1080 × 720.
10. special according to a kind of shelter bridge working condition detection system based on image processing techniques described in claim 5 or 6
Sign is: the picture number of the sample video image frame of the shelter bridge position sample database is no less than 10,000.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910489381.7A CN110210427B (en) | 2019-06-06 | 2019-06-06 | Corridor bridge working state detection system and method based on image processing technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910489381.7A CN110210427B (en) | 2019-06-06 | 2019-06-06 | Corridor bridge working state detection system and method based on image processing technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110210427A true CN110210427A (en) | 2019-09-06 |
| CN110210427B CN110210427B (en) | 2021-04-23 |
Family
ID=67791190
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910489381.7A Active CN110210427B (en) | 2019-06-06 | 2019-06-06 | Corridor bridge working state detection system and method based on image processing technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110210427B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112528729A (en) * | 2020-10-19 | 2021-03-19 | 浙江大华技术股份有限公司 | Video-based airplane bridge approach event detection method and device |
| CN112633049A (en) * | 2020-08-19 | 2021-04-09 | 中冶南方工程技术有限公司 | System for monitoring working state of blast furnace tuyere |
| CN113011335A (en) * | 2021-03-19 | 2021-06-22 | 捻果科技(深圳)有限公司 | Automatic identification method for corridor bridge approach connection aircraft in civil aviation airport flight area |
| CN115410418A (en) * | 2022-10-31 | 2022-11-29 | 北京千尧新能源科技开发有限公司 | Operation monitoring method and system for operation and maintenance boarding corridor bridge |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102921915A (en) * | 2012-10-23 | 2013-02-13 | 杭州谱诚泰迪实业有限公司 | Slag carry-over detection method and device based on image recognition of vortex on surface of molten steel |
| CN104268685A (en) * | 2014-09-23 | 2015-01-07 | 中国民航大学 | Airport ground vehicle tracking management system |
| CN104966045A (en) * | 2015-04-02 | 2015-10-07 | 北京天睿空间科技有限公司 | Video-based airplane entry-departure parking lot automatic detection method |
| CN105045179A (en) * | 2015-07-16 | 2015-11-11 | 中国民航大学 | Airport boarding bridge, ground air conditioning and power supply crew scheduling monitoring and management system |
| CN105116847A (en) * | 2015-07-16 | 2015-12-02 | 中国民航大学 | Airport boarding bridge, ground air-conditioning and power supply set operation scheduling monitoring system |
| CN107437245A (en) * | 2017-06-26 | 2017-12-05 | 西南交通大学 | High-speed railway touching net method for diagnosing faults based on depth convolutional neural networks |
| CN107705051A (en) * | 2017-11-16 | 2018-02-16 | 天津航佳机场设备科技有限公司 | A kind of airport seat in the plane equipment condition monitoring and metering settlement management system and method |
| CN108288272A (en) * | 2018-02-08 | 2018-07-17 | 首都师范大学 | Yarn recognition methods and device |
| US20190154442A1 (en) * | 2016-01-15 | 2019-05-23 | Fugro Roadware Inc. | High speed stereoscopic pavement surface scanning system and method |
-
2019
- 2019-06-06 CN CN201910489381.7A patent/CN110210427B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102921915A (en) * | 2012-10-23 | 2013-02-13 | 杭州谱诚泰迪实业有限公司 | Slag carry-over detection method and device based on image recognition of vortex on surface of molten steel |
| CN104268685A (en) * | 2014-09-23 | 2015-01-07 | 中国民航大学 | Airport ground vehicle tracking management system |
| CN104966045A (en) * | 2015-04-02 | 2015-10-07 | 北京天睿空间科技有限公司 | Video-based airplane entry-departure parking lot automatic detection method |
| CN105045179A (en) * | 2015-07-16 | 2015-11-11 | 中国民航大学 | Airport boarding bridge, ground air conditioning and power supply crew scheduling monitoring and management system |
| CN105116847A (en) * | 2015-07-16 | 2015-12-02 | 中国民航大学 | Airport boarding bridge, ground air-conditioning and power supply set operation scheduling monitoring system |
| US20190154442A1 (en) * | 2016-01-15 | 2019-05-23 | Fugro Roadware Inc. | High speed stereoscopic pavement surface scanning system and method |
| CN107437245A (en) * | 2017-06-26 | 2017-12-05 | 西南交通大学 | High-speed railway touching net method for diagnosing faults based on depth convolutional neural networks |
| CN107705051A (en) * | 2017-11-16 | 2018-02-16 | 天津航佳机场设备科技有限公司 | A kind of airport seat in the plane equipment condition monitoring and metering settlement management system and method |
| CN108288272A (en) * | 2018-02-08 | 2018-07-17 | 首都师范大学 | Yarn recognition methods and device |
Non-Patent Citations (3)
| Title |
|---|
| PENG ZHANG ET AL: "Airport Detection on Optical Satellite Images Using", 《IEEE GEOSCIENCE AND REMOTE SENSING LETTERS》 * |
| 张积洪等: "基于Linux 的嵌入式停机坪保障设备状态监测平台设计", 《数字技术与应用》 * |
| 赵鑫欣: "基于卷积神经网络的铁轨扣件异常状态检测研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112633049A (en) * | 2020-08-19 | 2021-04-09 | 中冶南方工程技术有限公司 | System for monitoring working state of blast furnace tuyere |
| CN112528729A (en) * | 2020-10-19 | 2021-03-19 | 浙江大华技术股份有限公司 | Video-based airplane bridge approach event detection method and device |
| CN113011335A (en) * | 2021-03-19 | 2021-06-22 | 捻果科技(深圳)有限公司 | Automatic identification method for corridor bridge approach connection aircraft in civil aviation airport flight area |
| CN115410418A (en) * | 2022-10-31 | 2022-11-29 | 北京千尧新能源科技开发有限公司 | Operation monitoring method and system for operation and maintenance boarding corridor bridge |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110210427B (en) | 2021-04-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110210427A (en) | A kind of shelter bridge working condition detection system and method based on image processing techniques | |
| CN106203265B (en) | A kind of Construction Fugitive Dust Pollution source monitors automatically and coverage forecasting system and method | |
| CN105373135B (en) | A kind of method and system of aircraft docking guidance and plane type recognition based on machine vision | |
| CN104966045B (en) | Aircraft disengaging berth automatic testing method based on video | |
| Lin et al. | Detection of buildings using perceptual grouping and shadows | |
| CN108109437B (en) | Unmanned aerial vehicle autonomous route extraction and generation method based on map features | |
| CN110047241A (en) | A kind of forest fire unmanned plane cruise monitoring system | |
| CN110261394A (en) | Online fan blade damages real-time diagnosis system and method | |
| CN109887343B (en) | Automatic acquisition and monitoring system and method for flight ground service support nodes | |
| CN106547814A (en) | A kind of power transmission line unmanned machine patrols and examines the structuring automatic archiving method of image | |
| CN104951775A (en) | Video technology based secure and smart recognition method for railway crossing protection zone | |
| CN106331636A (en) | Intelligent video monitoring system and method of oil pipelines based on behavioral event triggering | |
| CN113947731B (en) | Foreign matter identification method and system based on contact net safety inspection | |
| CN107181323B (en) | Unmanned aerial vehicle line patrol system and method | |
| CN106295695B (en) | A kind of takeoff and landing process automatic tracing image pickup method and device | |
| CN112162565B (en) | Uninterrupted self-main-pole tower inspection method based on multi-machine collaborative operation | |
| CN110046584B (en) | Road crack detection device and detection method based on unmanned aerial vehicle inspection | |
| CN108846331A (en) | The video frequency identifying method whether a kind of EMU chassis screw fastener falls off | |
| CN110070729A (en) | It is a kind of that vehicle detecting system and method are stopped based on the separated of mist calculating | |
| CN112270267A (en) | Camera shooting recognition system capable of automatically capturing line faults | |
| CN110211159A (en) | A kind of aircraft position detection system and method based on image/video processing technique | |
| CN115275870B (en) | Inspection system based on high-altitude line maintenance | |
| CN111651036A (en) | Unmanned aerial vehicle simulation training system and method for power inspection | |
| CN110363144A (en) | A kind of aircraft door switch state detecting system and method based on image processing techniques | |
| CN112613334A (en) | High-precision autonomous inspection image identification method of unmanned aerial vehicle on power transmission line |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |