CN110210427B - Corridor bridge working state detection system and method based on image processing technology - Google Patents
Corridor bridge working state detection system and method based on image processing technology Download PDFInfo
- Publication number
- CN110210427B CN110210427B CN201910489381.7A CN201910489381A CN110210427B CN 110210427 B CN110210427 B CN 110210427B CN 201910489381 A CN201910489381 A CN 201910489381A CN 110210427 B CN110210427 B CN 110210427B
- Authority
- CN
- China
- Prior art keywords
- bridge
- region
- gallery bridge
- gallery
- data
- 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
Images
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
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Signal Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a corridor bridge working state detection system and a corridor bridge working state detection method based on an image processing technology, and the corridor bridge working state detection system comprises a video acquisition system, a data acquisition system, a convolutional neural network system and a corridor bridge working state detection system, wherein the video acquisition system, the data acquisition system, the convolutional neural network system and the corridor bridge working state detection system are sequentially in communication connection, the data acquisition system comprises a data image frame electronic line dividing module and a data image frame acquisition module, a corridor bridge position sample database is arranged in the convolutional neural network system, and the corridor bridge working state detection system comprises a corridor bridge working state judgment module; the video acquisition system comprises a plurality of video cameras. The invention judges the current working condition of the gallery bridge by automatically analyzing the apron monitoring video signal, and can simultaneously obtain the starting time and the finishing time of approaching the gallery bridge to the airplane and the starting time and the finishing time of leaving the gallery bridge from the airplane.
Description
Technical Field
The invention relates to the technical field of gallery bridge working state detection, in particular to a gallery bridge working state detection system and method based on an image processing technology.
Background
The position of the corridor bridge is detected in the monitoring video of the airport flight area, and the starting time and the finishing time of the movement of the corridor bridge approaching the airplane and the starting time and the finishing time of the movement of the corridor bridge leaving the airplane in the flight guarantee process can be automatically extracted by combining the preset judgment logic, so that airport support staff can master the key process of flight guarantee. At present, an airport mainly judges the operation state of a gallery bridge by depending on a signal returned by a bridge-mounted system of the gallery bridge, but on one hand, the number of new gallery bridges for mounting the bridge-mounted system is small, the improvement cost is high, on the other hand, even if the bridge-mounted system is mounted, the possibility of equipment failure also exists, and the video detection can play a role in data backup.
Disclosure of Invention
Aiming at the defects in the prior art, the invention aims to provide a corridor bridge working state detection system and method based on an image processing technology, which can judge the current working condition of a corridor bridge by automatically analyzing a airport monitoring video signal and can simultaneously obtain the starting time and the finishing time of approaching the corridor bridge and the starting time and the finishing time of leaving the corridor bridge.
The purpose of the invention is realized by the following technical scheme:
a gallery bridge working state detection system based on an image processing technology comprises a video acquisition system, a data acquisition system, a convolutional neural network system and a gallery bridge working state detection system, wherein the video acquisition system, the data acquisition system, the convolutional neural network system and the gallery bridge working state detection system are sequentially in communication connection, the data acquisition system comprises a data image frame electronic wire dividing module and a data image frame acquisition module, a gallery bridge position sample database is arranged in the convolutional neural network system, and the gallery bridge working state detection system comprises a gallery bridge working state judgment module; the video acquisition system comprises a plurality of video cameras, the video acquisition system is used for acquiring video data and transmitting the video data to the data acquisition system in a combined manner, a data image frame acquisition module of the data acquisition system is used for acquiring video image frames one by one and transmitting the video image frames to a data image frame electronic line dividing module, the data image frame electronic line dividing module is used for dividing the video image frames into an electronic line A and an electronic line B, the electronic line A divides the video image frames into an A1 area and an A2 area, the electronic line B divides an A2 area of the video image frames into an A2B1 area and an A2B2 area, and the A2B1 area is close to the A1 area; the gallery bridge position sample database is used for storing sample data of the working state of the gallery bridge, wherein the sample data comprises sample data of starting to approach the bridge of the gallery bridge, sample data of finishing approaching the bridge of the gallery bridge, sample data of starting to leave the gallery bridge and sample data of finishing working of the gallery bridge; the convolutional neural network system is used for carrying out image feature extraction and comparison processing on video image frames acquired by the data acquisition system and sample data of the gallery and bridge position sample database, the gallery and bridge working state judgment module is used for obtaining the working state of the gallery bridge through image feature extraction, comparison processing and logic judgment according to the sample data of the gallery and bridge position sample database, and the working state of the gallery bridge comprises the steps that the gallery bridge starts to be close to the bridge, the gallery bridge is close to the bridge and is completed, the gallery bridge starts to leave and the gallery bridge finishes working.
In order to better realize the gallery bridge working state detection system, the convolutional neural network system is used for extracting the video image frames of the data acquisition system and identifying the gallery bridge characteristic region and the background characteristic region, and the gallery bridge characteristic region and the background characteristic region of the video image frames are transmitted to the gallery bridge working state detection system for comparison processing by the convolutional neural network system.
Preferably, the gallery bridge working state judgment module performs comparison processing by adopting an inter-frame difference method to obtain the gallery bridge position state.
Preferably, the convolutional neural network system inputs a new video image frame F at the current time ttFirstly, obtaining a binary black-and-white image F by an interframe difference methodFrame differenceBlack and white image FFrame differenceThe background feature area of (A) is uniformly black, and the black-and-white image F isFrame differenceThe aircraft feature areas of (a) are uniformly white.
A corridor bridge working state detection method based on an image processing technology comprises the following steps:
A. obtaining a gallery bridge position sample database: the method comprises the steps of collecting sample video data in a gallery bridge working area in an airport through a video collection system, wherein the sample video data comprises gallery bridge starting sample video data, gallery bridge stopping sample video data, gallery bridge starting sample video data and gallery bridge working ending sample video data, the video collection system transmits the sample video data to the data collection system, a data image frame collection module of the data collection system collects sample video image frames one by one on the sample video data and transmits the sample video image frames to a data image frame electronic line dividing module, the data image frame electronic line dividing module divides a sample video image frame into an electronic line A and an electronic line B, the electron beam A divides the sample video image frame into an A1 region and an A2 region, the electron beam B divides the A2 region of the sample video image frame into an A2B1 region and an A2B2 region, and the A2B1 region is close to the A1 region; the convolutional neural network system extracts and identifies sample video image frames to obtain a gallery bridge characteristic region and a background characteristic region, and stores the sample video image frames in a gallery bridge position sample database after binarization processing and feature unification;
B. the method comprises the steps that real-time video data in a working area of a gallery bridge in an airport are collected through a video collection system, the video collection system transmits the real-time video data to the data collection system, a data image frame collection module of the data collection system collects real-time video image frames one by one and transmits the real-time video image frames to a data image frame electronic line dividing module, the data image frame electronic line dividing module divides the real-time video image frames into an electronic line A and an electronic line B, the electronic line A divides the real-time video image frames into an A1 area and an A2 area, the electronic line B divides an A2 area of the real-time video image frames into an A2B1 area and an A2B2 area, and the A2B1 area is close to an A1 area; the convolutional neural network system extracts and identifies real-time video image frames to obtain a gallery bridge feature region and a background feature region, and performs binarization processing and feature unification on the real-time video image frames, and then performs image feature extraction and comparison processing on the real-time video image frames and sample video image frames stored in a gallery bridge position sample database;
C. the corridor bridge working state data are obtained, and the corridor bridge working state judgment module is used for obtaining the corridor bridge working state through image feature extraction, comparison processing and logic judgment according to the real-time video data, wherein the corridor bridge working state comprises the corridor bridge starting to approach the corridor bridge, the corridor bridge approach the corridor bridge finishing, the corridor bridge starting to leave and the corridor bridge finishing; the corridor bridge working state judgment method of the corridor bridge working state judgment module comprises the following steps:
c1, if the center line of the gallery bridge feature region of the continuous M-second real-time video image frames is in the A1 region, judging that the gallery bridge does not work;
c2, if the center line of the gallery bridge feature region of the real-time video image frame enters the A2B1 region from the A1 region and the center line of the gallery bridge feature region of the continuous N-second real-time video image frame stays in the A2B1 region, judging that the operating state of the gallery bridge is that the gallery bridge starts to lean against the bridge;
c3, if the center line of the gallery bridge feature region of the real-time video image frame enters the A2B2 region from the A2B1 region and the center line of the gallery bridge feature region of the continuous N-second real-time video image frame stays in the A2B2 region, judging that the working state of the gallery bridge is that the gallery bridge is closed;
c4, if the center line of the gallery bridge feature region of the real-time video image frame enters the A2B1 region from the A2B2 region and the center line of the gallery bridge feature region of the continuous N-second real-time video image frame stays in the A2B1 region, judging that the operating state of the gallery bridge is that the gallery bridge starts to leave;
and C5, if the center line of the gallery bridge feature region of the real-time video image frame enters the A1 region from the A2B1 region and the center line of the gallery bridge feature region of the continuous N-second real-time video image frame stays in the A1 region, judging that the working state of the gallery bridge is the end of the gallery bridge.
A corridor bridge working state detection method based on an image processing technology comprises the following steps:
A. obtaining a gallery bridge position sample database: the method comprises the steps of collecting sample video data in a gallery bridge working area in an airport through a video collection system, wherein the sample video data comprises gallery bridge starting sample video data, gallery bridge stopping sample video data, gallery bridge starting sample video data and gallery bridge working ending sample video data, the video collection system transmits the sample video data to the data collection system, a data image frame collection module of the data collection system collects sample video image frames one by one on the sample video data and transmits the sample video image frames to a data image frame electronic line dividing module, the data image frame electronic line dividing module divides a sample video image frame into an electronic line A and an electronic line B, the electron beam A divides the sample video image frame into an A1 region and an A2 region, the electron beam B divides the A2 region of the sample video image frame into an A2B1 region and an A2B2 region, and the A2B1 region is close to the A1 region; the method comprises the following steps that a convolutional neural network system extracts and identifies sample video image frames to obtain a gallery bridge characteristic region and a background characteristic region, the sample video image frames are subjected to binarization processing and feature unification and then are transmitted to a gallery bridge working state judgment module, and the gallery bridge working state judgment module obtains the working state of a gallery bridge through image feature extraction, comparison processing and logic judgment and transmits and stores the working state of the gallery bridge in an airplane position sample database; the method for judging the gallery bridge working state of the sample video image frame by the gallery bridge working state judging module comprises the following steps:
a1, if the center line of the gallery bridge feature region of the continuous M-second sample video image frames is in the A1 region, judging that the gallery bridge does not work;
a2, if the center line of the gallery bridge feature region of the sample video image frame enters the A2B1 region from the A1 region and the center line of the gallery bridge feature region of the sample video image frame for N seconds continuously stays in the A2B1 region, judging that the working state of the gallery bridge is that the gallery bridge starts to approach the bridge, and correspondingly storing the sample video image frame at the position as gallery bridge starting to approach the bridge sample video data of a gallery bridge position sample database;
a3, if the center line of the gallery bridge feature region of the sample video image frame enters the A2B2 region from the A2B1 region and the center line of the gallery bridge feature region of the sample video image frame for N seconds continuously stays in the A2B2 region, judging that the working state of the gallery bridge is that the gallery bridge is closed, and then correspondingly storing the sample video image frame at the position as gallery bridge closing bridge finishing sample video data of a gallery bridge position sample database;
a4, if the center line of the gallery bridge feature region of the sample video image frame enters the A2B1 region from the A2B2 region and the center line of the gallery bridge feature region of the sample video image frame for N seconds continuously stays in the A2B1 region, judging that the gallery bridge starts to leave in the working state, and correspondingly storing the sample video image frame at the working state as the gallery bridge of the gallery bridge position sample database and starting to leave the sample video data;
a5, if the center line of the gallery bridge feature region of the sample video image frame enters the A1 region from the A2B1 region and the center line of the gallery bridge feature region of the sample video image frame for N seconds continuously stays in the A1 region, judging that the working state of the gallery bridge is the gallery bridge working end, and correspondingly storing the sample video image frame at the position as gallery bridge working end sample video data of a gallery bridge position sample database;
B. the method comprises the steps that real-time video data in a working area of a gallery bridge in an airport are collected through a video collection system, the video collection system transmits the real-time video data to the data collection system, a data image frame collection module of the data collection system collects real-time video image frames one by one and transmits the real-time video image frames to a data image frame electronic line dividing module, the data image frame electronic line dividing module divides the real-time video image frames into an electronic line A and an electronic line B, the electronic line A divides the real-time video image frames into an A1 area and an A2 area, the electronic line B divides an A2 area of the real-time video image frames into an A2B1 area and an A2B2 area, and the A2B1 area is close to an A1 area; the convolutional neural network system extracts and identifies real-time video image frames to obtain a gallery bridge feature region and a background feature region, and performs binarization processing, image feature extraction and feature unification on the real-time video image frames and then performs feature comparison processing on the real-time video image frames and sample video image frames stored in a gallery bridge position sample database;
C. the corridor bridge working state data are obtained, and the corridor bridge working state judgment module is used for obtaining the corridor bridge working state through image feature extraction, comparison processing and logic judgment according to the real-time video data, wherein the corridor bridge working state comprises the corridor bridge starting to approach the corridor bridge, the corridor bridge approach the corridor bridge finishing, the corridor bridge starting to leave and the corridor bridge finishing; the corridor bridge working state judgment method of the corridor bridge working state judgment module comprises the following steps:
c1, if the center line of the gallery bridge feature region of the continuous M-second real-time video image frames is in the A1 region, judging that the gallery bridge does not work;
c2, if the center line of the gallery bridge feature region of the real-time video image frame enters the A2B1 region from the A1 region and the center line of the gallery bridge feature region of the continuous N-second real-time video image frame stays in the A2B1 region, judging that the operating state of the gallery bridge is that the gallery bridge starts to lean against the bridge;
c3, if the center line of the gallery bridge feature region of the real-time video image frame enters the A2B2 region from the A2B1 region and the center line of the gallery bridge feature region of the continuous N-second real-time video image frame stays in the A2B2 region, judging that the working state of the gallery bridge is that the gallery bridge is closed;
c4, if the center line of the gallery bridge feature region of the real-time video image frame enters the A2B1 region from the A2B2 region and the center line of the gallery bridge feature region of the continuous N-second real-time video image frame stays in the A2B1 region, judging that the operating state of the gallery bridge is that the gallery bridge starts to leave;
and C5, if the center line of the gallery bridge feature region of the real-time video image frame enters the A1 region from the A2B1 region and the center line of the gallery bridge feature region of the continuous N-second real-time video image frame stays in the A1 region, judging that the working state of the gallery bridge is the end of the gallery bridge.
Preferably, M in step C is an integer of 10 or more, and N in step C is an integer of 5 to 10.
Preferably, the sample video data in the step a includes gallery and bridge position sample data in night, rain and snow, fog or haze and sand weather, wherein the gallery and bridge position sample data in the night weather is not less than 10%, wherein the gallery and bridge position sample data in the rain and snow weather is not less than 10%, wherein the gallery and bridge position sample data in the fog or haze weather is not less than 10%, and wherein the gallery and bridge position sample data in the sand weather is not less than 10%.
Preferably, the video acquisition system in the step A and the step B comprises a plurality of video cameras, and all the video cameras are respectively arranged at the positions right in front, left in front, right above, left above and back above the corridor bridge working area in the airport; the data acquisition system in the step A and the data acquisition system in the step B acquire single video data of all the video cameras, combine the single video data into video data and transmit the video data to the data image frame electronic line dividing module; the video camera is a digital camera, and the resolution of the digital camera is not lower than 1080 multiplied by 720.
Preferably, the number of the pictures of the sample video image frames of the gallery bridge position sample database is not less than ten thousand.
Compared with the prior art, the invention has the following advantages and beneficial effects:
the invention judges the current working condition of the gallery bridge by automatically analyzing the airport monitoring video signal, and can simultaneously obtain the starting time and the finishing time of approaching the airplane to the gallery bridge and the starting time and the finishing time of leaving the airplane from the gallery bridge, so that the control personnel can continuously guide the airplane on the scene, and an accurate time basis is provided for the airport flight guarantee.
Drawings
Fig. 1 is a schematic block diagram of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the following examples:
example one
As shown in fig. 1, a gallery bridge working state detection system based on an image processing technology comprises a video acquisition system, a data acquisition system, a convolutional neural network system and a gallery bridge working state detection system, wherein the video acquisition system, the data acquisition system, the convolutional neural network system and the gallery bridge working state detection system are sequentially in communication connection, the data acquisition system comprises a data image frame electronic line dividing module and a data image frame acquisition module, a gallery bridge position sample database is arranged in the convolutional neural network system, and the gallery bridge working state detection system comprises a gallery bridge working state judgment module; the video acquisition system comprises a plurality of video cameras, the video acquisition system is used for acquiring video data and transmitting the video data to the data acquisition system in a combined manner, a data image frame acquisition module of the data acquisition system is used for acquiring video image frames one by one and transmitting the video image frames to a data image frame electronic line dividing module, the data image frame electronic line dividing module is used for dividing the video image frames into an electronic line A and an electronic line B, the electronic line A divides the video image frames into an A1 area and an A2 area, the electronic line B divides an A2 area of the video image frames into an A2B1 area and an A2B2 area, and the A2B1 area is close to the A1 area; the gallery bridge position sample database is used for storing sample data of the working state of the gallery bridge, wherein the sample data comprises sample data of starting to approach the bridge of the gallery bridge, sample data of finishing approaching the bridge of the gallery bridge, sample data of starting to leave the gallery bridge and sample data of finishing working of the gallery bridge; the convolutional neural network system is used for carrying out image feature extraction and comparison processing on video image frames acquired by the data acquisition system and sample data of the gallery and bridge position sample database, the gallery and bridge working state judgment module is used for obtaining the working state of the gallery bridge through image feature extraction, comparison processing and logic judgment according to the sample data of the gallery and bridge position sample database, and the working state of the gallery bridge comprises the steps that the gallery bridge starts to be close to the bridge, the gallery bridge is close to the bridge and is completed, the gallery bridge starts to leave and the gallery bridge finishes working.
The convolutional neural network system is used for extracting video image frames of the data acquisition system and identifying a gallery bridge characteristic region and a background characteristic region, and the gallery bridge characteristic region and the background characteristic region of the video image frames are transmitted to the gallery bridge working state detection system for comparison processing by the convolutional neural network system.
And the gallery bridge working state judgment module performs comparison processing by adopting an inter-frame difference method and obtains the gallery bridge position state.
The convolution neural network system inputs a new video image frame F at the current moment ttFirstly, obtaining a binary black-and-white image F by an interframe difference methodFrame differenceBlack and white image FFrame differenceThe background feature area of (A) is uniformly black, and the black-and-white image F isFrame differenceThe aircraft feature areas of (a) are uniformly white.
A corridor bridge working state detection method based on an image processing technology comprises the following steps:
A. obtaining a gallery bridge position sample database: the method comprises the steps of collecting sample video data in a gallery bridge working area in an airport through a video collection system, wherein the sample video data comprises gallery bridge starting sample video data, gallery bridge stopping sample video data, gallery bridge starting sample video data and gallery bridge working ending sample video data, the video collection system transmits the sample video data to the data collection system, a data image frame collection module of the data collection system collects sample video image frames one by one on the sample video data and transmits the sample video image frames to a data image frame electronic line dividing module, the data image frame electronic line dividing module divides a sample video image frame into an electronic line A and an electronic line B, the electron beam A divides the sample video image frame into an A1 region and an A2 region, the electron beam B divides the A2 region of the sample video image frame into an A2B1 region and an A2B2 region, and the A2B1 region is close to the A1 region; the convolutional neural network system extracts and identifies sample video image frames to obtain a gallery bridge characteristic region and a background characteristic region, and stores the sample video image frames in a gallery bridge position sample database after binarization processing and feature unification; the gallery bridge starting bridge approaching sample video data of the gallery bridge position sample database in the embodiment directly come from sample video data determined when the gallery bridge starts to approach the bridge (the sample video data is sample video data acquired by a video acquisition system when the gallery bridge starts to approach the bridge); the gallery bridge approach completion sample video data of the gallery bridge position sample database in the embodiment directly come from the sample video data determined as the gallery bridge approach completion (the sample video data is the sample video data acquired by the video acquisition system when the gallery bridge approach completion is determined); the gallery bridge starting leaving sample video data of the gallery bridge position sample database in the embodiment are directly derived from sample video data which is determined as the gallery bridge starting leaving (the sample video data is the sample video data which is acquired by the video acquisition system when the gallery bridge starting leaving is determined); the gallery bridge working end sample video data of the gallery bridge position sample database in the embodiment is directly derived from the sample video data determined as the gallery bridge working end (the sample video data is the sample video data acquired by the video acquisition system when the gallery bridge working end is determined). The preferable sample video data comprises gallery and bridge position sample data under the weather conditions of night, rain, snow, fog or haze and sand dust, wherein the gallery and bridge position sample data under the weather conditions of night is not less than 10%, the gallery and bridge position sample data under the weather conditions of rain and snow is not less than 10%, the gallery and bridge position sample data under the weather conditions of fog or haze is not less than 10%, and the gallery and bridge position sample data under the weather conditions of sand dust is not less than 10%. The number of the pictures of the sample video image frames of the gallery bridge position sample database is not less than ten thousand.
B. The method comprises the steps that real-time video data in a working area of a gallery bridge in an airport are collected through a video collection system, the video collection system transmits the real-time video data to the data collection system, a data image frame collection module of the data collection system collects real-time video image frames one by one and transmits the real-time video image frames to a data image frame electronic line dividing module, the data image frame electronic line dividing module divides the real-time video image frames into an electronic line A and an electronic line B, the electronic line A divides the real-time video image frames into an A1 area and an A2 area, the electronic line B divides an A2 area of the real-time video image frames into an A2B1 area and an A2B2 area, and the A2B1 area is close to an A1 area; the convolutional neural network system extracts and identifies the real-time video image frames to obtain a gallery bridge feature region and a background feature region, and performs binarization processing and feature unification on the real-time video image frames, and then performs image feature extraction and comparison processing on the real-time video image frames and sample video image frames stored in a gallery bridge position sample database.
C. The corridor bridge working state data are obtained, and the corridor bridge working state judgment module is used for obtaining the corridor bridge working state through image feature extraction, comparison processing and logic judgment according to the real-time video data, wherein the corridor bridge working state comprises the corridor bridge starting to approach the corridor bridge, the corridor bridge approach the corridor bridge finishing, the corridor bridge starting to leave and the corridor bridge finishing; the corridor bridge working state judgment method of the corridor bridge working state judgment module comprises the following steps:
c1, if the center line of the gallery bridge feature region of the continuous M-second real-time video image frames is in the A1 region, judging that the gallery bridge does not work;
c2, if the center line of the gallery bridge feature region of the real-time video image frame enters the A2B1 region from the A1 region and the center line of the gallery bridge feature region of the continuous N-second real-time video image frame stays in the A2B1 region, judging that the operating state of the gallery bridge is that the gallery bridge starts to lean against the bridge;
c3, if the center line of the gallery bridge feature region of the real-time video image frame enters the A2B2 region from the A2B1 region and the center line of the gallery bridge feature region of the continuous N-second real-time video image frame stays in the A2B2 region, judging that the working state of the gallery bridge is that the gallery bridge is closed;
c4, if the center line of the gallery bridge feature region of the real-time video image frame enters the A2B1 region from the A2B2 region and the center line of the gallery bridge feature region of the continuous N-second real-time video image frame stays in the A2B1 region, judging that the operating state of the gallery bridge is that the gallery bridge starts to leave;
and C5, if the center line of the gallery bridge feature region of the real-time video image frame enters the A1 region from the A2B1 region and the center line of the gallery bridge feature region of the continuous N-second real-time video image frame stays in the A1 region, judging that the working state of the gallery bridge is the end of the gallery bridge.
In step C, M is an integer of more than 10, and N in step C is an integer between 5 and 10.
The video acquisition system in the steps A and B comprises a plurality of video cameras, wherein all the video cameras are respectively arranged at the positions right in front, left in front, right above, left above and back above the corridor bridge working area in the airport; the data acquisition system in the step A and the data acquisition system in the step B acquire single video data of all the video cameras, combine the single video data into video data and transmit the video data to the data image frame electronic line dividing module; the video camera is a digital camera, and the resolution of the digital camera is not lower than 1080 multiplied by 720.
Example two
A corridor bridge working state detection method based on an image processing technology comprises the following steps:
A. obtaining a gallery bridge position sample database: the method comprises the steps of collecting sample video data in a gallery bridge working area in an airport through a video collection system, wherein the sample video data comprises gallery bridge starting sample video data, gallery bridge stopping sample video data, gallery bridge starting sample video data and gallery bridge working ending sample video data, the video collection system transmits the sample video data to the data collection system, a data image frame collection module of the data collection system collects sample video image frames one by one on the sample video data and transmits the sample video image frames to a data image frame electronic line dividing module, the data image frame electronic line dividing module divides a sample video image frame into an electronic line A and an electronic line B, the electron beam A divides the sample video image frame into an A1 region and an A2 region, the electron beam B divides the A2 region of the sample video image frame into an A2B1 region and an A2B2 region, and the A2B1 region is close to the A1 region; the method comprises the following steps that a convolutional neural network system extracts and identifies sample video image frames to obtain a gallery bridge characteristic region and a background characteristic region, the sample video image frames are subjected to binarization processing and feature unification and then are transmitted to a gallery bridge working state judgment module, and the gallery bridge working state judgment module obtains the working state of a gallery bridge through image feature extraction, comparison processing and logic judgment and transmits and stores the working state of the gallery bridge in an airplane position sample database; the gallery bridge starting bridge approaching sample video data of the gallery bridge position sample database can be derived from sample video data when the gallery bridge working state judgment module judges that the gallery bridge starts to approach the bridge, and can also be derived from sample video data determined as when the gallery bridge starts to approach the bridge; the gallery bridge approach completion sample video data of the gallery bridge position sample database can be derived from sample video data determined by the gallery bridge working state judgment module to be complete when the gallery bridge approach is completed, and can also be derived from sample video data determined to be complete when the gallery bridge approach is completed; the gallery bridge starting leaving sample video data of the gallery bridge position sample database can be derived from sample video data determined by the gallery bridge working state judgment module as the gallery bridge starting leaving, and can also be derived from sample video data determined as the gallery bridge starting leaving; the gallery bridge working end sample video data of the gallery bridge position sample database in this embodiment may be derived from the sample video data determined by the gallery bridge working state determination module as the gallery bridge working end, or may be derived from the sample video data determined as the gallery bridge working end. The method for judging the gallery bridge working state of the sample video image frame by the gallery bridge working state judging module comprises the following steps:
a1, if the center line of the gallery bridge feature region of the continuous M-second sample video image frames is in the A1 region, judging that the gallery bridge does not work;
a2, if the center line of the gallery bridge feature region of the sample video image frame enters the A2B1 region from the A1 region and the center line of the gallery bridge feature region of the sample video image frame for N seconds continuously stays in the A2B1 region, judging that the working state of the gallery bridge is that the gallery bridge starts to approach the bridge, and correspondingly storing the sample video image frame at the position as gallery bridge starting to approach the bridge sample video data of a gallery bridge position sample database;
a3, if the center line of the gallery bridge feature region of the sample video image frame enters the A2B2 region from the A2B1 region and the center line of the gallery bridge feature region of the sample video image frame for N seconds continuously stays in the A2B2 region, judging that the working state of the gallery bridge is that the gallery bridge is closed, and then correspondingly storing the sample video image frame at the position as gallery bridge closing bridge finishing sample video data of a gallery bridge position sample database;
a4, if the center line of the gallery bridge feature region of the sample video image frame enters the A2B1 region from the A2B2 region and the center line of the gallery bridge feature region of the sample video image frame for N seconds continuously stays in the A2B1 region, judging that the gallery bridge starts to leave in the working state, and correspondingly storing the sample video image frame at the working state as the gallery bridge of the gallery bridge position sample database and starting to leave the sample video data;
a5, if the center line of the gallery bridge feature region of the sample video image frame enters the A1 region from the A2B1 region and the center line of the gallery bridge feature region of the sample video image frame for N seconds continuously stays in the A1 region, judging that the working state of the gallery bridge is the gallery bridge working end, and correspondingly storing the sample video image frame at the position as gallery bridge working end sample video data of a gallery bridge position sample database;
the sample video data comprises gallery and bridge position sample data under the weather conditions of night, rain, snow, fog or haze and sand dust, wherein the gallery and bridge position sample data under the weather conditions of night is not less than 10%, the gallery and bridge position sample data under the weather conditions of rain and snow is not less than 10%, the gallery and bridge position sample data under the weather conditions of fog or haze is not less than 10%, and the gallery and bridge position sample data under the weather conditions of sand dust is not less than 10%. The number of the pictures of the sample video image frames of the gallery bridge position sample database is not less than ten thousand.
B. The method comprises the steps that real-time video data in a working area of a gallery bridge in an airport are collected through a video collection system, the video collection system transmits the real-time video data to the data collection system, a data image frame collection module of the data collection system collects real-time video image frames one by one and transmits the real-time video image frames to a data image frame electronic line dividing module, the data image frame electronic line dividing module divides the real-time video image frames into an electronic line A and an electronic line B, the electronic line A divides the real-time video image frames into an A1 area and an A2 area, the electronic line B divides an A2 area of the real-time video image frames into an A2B1 area and an A2B2 area, and the A2B1 area is close to an A1 area; the convolutional neural network system extracts and identifies the real-time video image frames to obtain a gallery bridge feature region and a background feature region, performs binarization processing, image feature extraction and feature unification on the real-time video image frames, and then performs feature comparison processing on the real-time video image frames and sample video image frames stored in a gallery bridge position sample database, and the convolutional neural network system compares the real-time video image frames with the sample video image frames to obtain an initial working state conclusion of a gallery bridge.
C. The corridor bridge working state data are obtained, and the corridor bridge working state judgment module is used for obtaining the corridor bridge working state through image feature extraction, comparison processing and logic judgment according to the real-time video data, wherein the corridor bridge working state comprises the corridor bridge starting to approach the corridor bridge, the corridor bridge approach the corridor bridge finishing, the corridor bridge starting to leave and the corridor bridge finishing; the corridor bridge working state judgment method of the corridor bridge working state judgment module comprises the following steps:
c1, if the center line of the gallery bridge feature region of the continuous M-second real-time video image frames is in the A1 region, judging that the gallery bridge does not work;
c2, if the center line of the gallery bridge feature region of the real-time video image frame enters the A2B1 region from the A1 region and the center line of the gallery bridge feature region of the continuous N-second real-time video image frame stays in the A2B1 region, judging that the operating state of the gallery bridge is that the gallery bridge starts to lean against the bridge;
c3, if the center line of the gallery bridge feature region of the real-time video image frame enters the A2B2 region from the A2B1 region and the center line of the gallery bridge feature region of the continuous N-second real-time video image frame stays in the A2B2 region, judging that the working state of the gallery bridge is that the gallery bridge is closed;
c4, if the center line of the gallery bridge feature region of the real-time video image frame enters the A2B1 region from the A2B2 region and the center line of the gallery bridge feature region of the continuous N-second real-time video image frame stays in the A2B1 region, judging that the operating state of the gallery bridge is that the gallery bridge starts to leave;
and C5, if the center line of the gallery bridge feature region of the real-time video image frame enters the A1 region from the A2B1 region and the center line of the gallery bridge feature region of the continuous N-second real-time video image frame stays in the A1 region, judging that the working state of the gallery bridge is the end of the gallery bridge.
M in the step C is an integer of more than 10, and N in the step C is an integer between 5 and 10.
The corridor bridge working state detection system provided by the invention is used for logically judging the corridor bridge working state and finally obtaining corridor bridge working state data by combining the preliminary working state conclusion.
The video acquisition system in the steps A and B comprises a plurality of video cameras, wherein all the video cameras are respectively arranged at the positions right in front, left in front, right above, left above and back above the corridor bridge working area in the airport; the data acquisition system in the step A and the data acquisition system in the step B acquire single video data of all the video cameras, combine the single video data into video data and transmit the video data to the data image frame electronic line dividing module; the video camera is a digital camera, and the resolution of the digital camera is not lower than 1080 multiplied by 720.
EXAMPLE III
A corridor bridge working state detection method based on an image processing technology comprises the following steps:
A. and establishing a gallery bridge picture sample library for neural network training.
B. And training a convolutional neural network for corridor bridge recognition, which is hereinafter referred to as a recognition network.
C. And constructing a gallery bridge detection model frame, which is hereinafter referred to as a detection frame.
D. And connecting the detection frame with a data acquisition module which is responsible for collecting monitoring videos.
E. The data acquisition module transfers the video data to the detection framework frame by frame.
F. The detection frame processes each input frame of video image, and finally generates a corridor bridge detection result, namely a corridor bridge detection frame.
And judging the working state of the gallery bridge according to the position of the gallery bridge detection frame, wherein the working state of the gallery bridge comprises non-working, starting to approach the bridge (namely starting to approach the airplane), finishing approaching the bridge, starting to remove the bridge (namely starting to leave the airplane) and finishing removing the bridge.
Optionally, the process for judging the working state of the gallery bridge includes the following steps:
an electronic line A is divided in the monitoring video, and the A divides the apron monitoring image into two areas A1 and A2. When the gallery bridge is in a non-operating state, the central point of the gallery bridge detection frame is always located in the area A1.
The division of the electronic line B in the surveillance video divides the A2 area into an A2B1 area and an A2B2 area.
And if the center point of the detection result frame of the gallery bridge enters the A2B1 area from the A1 area and stays in the A2 area for N seconds continuously, setting the working state of the gallery bridge as the gallery bridge to approach the bridge.
And if the gallery bridge in the working state enters the A2B2 area from the A2B1 area and stays in the A2B2 area for N seconds continuously, the gallery bridge is considered to be completed by the bridge.
If the corridor bridge in the on-board state enters the A2B1 region from the A2B2 region and stays in the A2B1 region for N seconds, the corridor bridge is considered to be finished leaving the airplane.
If the gallery bridge enters the A1 area from the A2B1 area and stays in the A1 area for N seconds continuously, the gallery bridge work is considered to be finished.
The value of N in this embodiment is preferably an integer between 5 and 10.
And recording and outputting the working state of the corridor bridge to the outside.
The gallery bridge picture sample library establishing process comprises the following steps:
and acquiring a certain number of pictures containing the gallery bridge from the airport apron monitoring video, and labeling the pictures. Optionally, the number of the pictures in the sample library is not less than ten thousand. Optionally, the color picture collected by the digital camera is selected, and the resolution of the digital camera is not lower than 1080 × 720. Optionally, in order to ensure that the corridor bridge can be stably identified by the system under different viewing angles, pictures are respectively collected from the video monitoring devices installed right in front of the airport apron, in the upper left of the airport apron and in the upper back of the airport apron, and the number of the pictures collected by different devices is the same. Preferably, in order to ensure the adaptability of the neural network to weather such as night, rain, snow, fog, haze and dust, a part of the selected pictures should be collected under the above weather conditions. Preferably, the pictures collected in the scenes of night, rainy day, snowy day, fog day or haze day should respectively account for more than 10% of the picture sample library.
The object detection convolutional neural network training process of the embodiment comprises the following steps:
and training the object detection convolutional neural network by using the labeled training sample. Preferably, the sample labeling method and neural network training method can be trained according to Redmon J, Divvala S K, Girshick R B, et al, you Only Look Once in Unifield, Real-Time Object Detection [ J ] computer vision and pattern recognition,2016: 779-.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents and improvements made within the spirit and principle of the present invention are intended to be included within the scope of the present invention.
Claims (7)
1. The utility model provides a corridor bridge operating condition detecting system based on image processing technique which characterized in that: the system comprises a video acquisition system, a data acquisition system, a convolutional neural network system and a corridor bridge working state detection system, wherein the video acquisition system, the data acquisition system, the convolutional neural network system and the corridor bridge working state detection system are sequentially in communication connection, the data acquisition system comprises a data image frame electronic wire dividing module and a data image frame acquisition module, a corridor bridge position sample database is arranged in the convolutional neural network system, and the corridor bridge working state detection system comprises a corridor bridge working state judgment module; the video acquisition system comprises a plurality of video cameras, the video acquisition system is used for acquiring video data and transmitting the video data to the data acquisition system in a combined manner, a data image frame acquisition module of the data acquisition system is used for acquiring video image frames one by one and transmitting the video image frames to a data image frame electronic line dividing module, the data image frame electronic line dividing module is used for dividing the video image frames into an electronic line A and an electronic line B, the electronic line A divides the video image frames into an A1 area and an A2 area, the electronic line B divides an A2 area of the video image frames into an A2B1 area and an A2B2 area, and the A2B1 area is close to the A1 area; the gallery bridge position sample database is used for storing sample data of the working state of the gallery bridge, wherein the sample data comprises sample data of starting to approach the bridge of the gallery bridge, sample data of finishing approaching the bridge of the gallery bridge, sample data of starting to leave the gallery bridge and sample data of finishing working of the gallery bridge; the convolutional neural network system is used for carrying out image feature extraction and comparison processing on the video image frames acquired by the data acquisition system and sample data of the gallery bridge position sample database, and comparing the real-time video image frames with the sample video image frames to obtain a preliminary working state conclusion of the gallery bridge; the gallery bridge working state detection system combines the preliminary working state conclusion to judge and process through the position relation logics of the gallery bridge characteristic region central lines of the real-time video image frames in the A1 area, the A2B1 area and the A2B2 area and finally obtains the gallery bridge working state, wherein the gallery bridge working state comprises the beginning of approaching the bridge, the completion of approaching the bridge, the beginning of leaving of the bridge and the ending of working of the bridge.
2. The gallery bridge working state detection system based on image processing technology as claimed in claim 1, wherein: the convolutional neural network system adopts the following method to carry out video image frame FtAnd (3) processing: the convolutional neural network system inputs a new video image frame F at the current moment ttFirstly, obtaining a binary black-and-white image F by an interframe difference methodFrame differenceBlack and white image FFrame differenceThe background feature area of (A) is uniformly black, and the black-and-white image F isFrame differenceThe gallery bridge feature areas are uniformly white.
3. A corridor bridge working state detection method based on an image processing technology is characterized by comprising the following steps: the method comprises the following steps:
A. obtaining a gallery bridge position sample database: the method comprises the steps of collecting sample video data in a gallery bridge working area in an airport through a video collection system, wherein the sample video data comprises gallery bridge starting sample video data, gallery bridge stopping sample video data, gallery bridge starting sample video data and gallery bridge working ending sample video data, the video collection system transmits the sample video data to the data collection system, a data image frame collection module of the data collection system collects sample video image frames one by one on the sample video data and transmits the sample video image frames to a data image frame electronic line dividing module, the data image frame electronic line dividing module divides a sample video image frame into an electronic line A and an electronic line B, the electron beam A divides the sample video image frame into an A1 region and an A2 region, the electron beam B divides the A2 region of the sample video image frame into an A2B1 region and an A2B2 region, and the A2B1 region is close to the A1 region; the convolutional neural network system extracts and identifies sample video image frames to obtain a gallery bridge characteristic region and a background characteristic region, and stores the sample video image frames in a gallery bridge position sample database after binarization processing and feature unification;
B. the method comprises the steps that real-time video data in a working area of a gallery bridge in an airport are collected through a video collection system, the video collection system transmits the real-time video data to the data collection system, a data image frame collection module of the data collection system collects real-time video image frames one by one and transmits the real-time video image frames to a data image frame electronic line dividing module, the data image frame electronic line dividing module divides the real-time video image frames into an electronic line A and an electronic line B, the electronic line A divides the real-time video image frames into an A1 area and an A2 area, the electronic line B divides an A2 area of the real-time video image frames into an A2B1 area and an A2B2 area, and the A2B1 area is close to an A1 area; the convolutional neural network system extracts and identifies real-time video image frames to obtain a gallery bridge characteristic region and a background characteristic region, and compares the real-time video image frames with sample video image frames stored in a gallery bridge position sample database after binarization processing and characteristic unification and obtains a preliminary working state conclusion of a gallery bridge;
C. working state data of the gallery bridge are obtained, and the working state of the gallery bridge is obtained through logic judgment according to a preliminary working state conclusion of the gallery bridge by a gallery bridge working state judgment module, wherein the working state of the gallery bridge comprises the beginning of bridge abutment of the gallery bridge, the completion of bridge abutment of the gallery bridge, the beginning of departure of the gallery bridge and the end of the work of the gallery bridge; the corridor bridge working state judgment method of the corridor bridge working state judgment module comprises the following steps:
c1, if the center line of the gallery bridge feature region of the continuous M-second real-time video image frames is in the A1 region, judging that the gallery bridge does not work;
c2, if the center line of the gallery bridge feature region of the real-time video image frame enters the A2B1 region from the A1 region and the center line of the gallery bridge feature region of the continuous N-second real-time video image frame stays in the A2B1 region, judging that the operating state of the gallery bridge is that the gallery bridge starts to lean against the bridge;
c3, if the center line of the gallery bridge feature region of the real-time video image frame enters the A2B2 region from the A2B1 region and the center line of the gallery bridge feature region of the continuous N-second real-time video image frame stays in the A2B2 region, judging that the working state of the gallery bridge is that the gallery bridge is closed;
c4, if the center line of the gallery bridge feature region of the real-time video image frame enters the A2B1 region from the A2B2 region and the center line of the gallery bridge feature region of the continuous N-second real-time video image frame stays in the A2B1 region, judging that the operating state of the gallery bridge is that the gallery bridge starts to leave;
and C5, if the center line of the gallery bridge feature region of the real-time video image frame enters the A1 region from the A2B1 region and the center line of the gallery bridge feature region of the continuous N-second real-time video image frame stays in the A1 region, judging that the working state of the gallery bridge is the end of the gallery bridge.
4. A corridor bridge working state detection method based on an image processing technology is characterized by comprising the following steps: the method comprises the following steps:
A. obtaining a gallery bridge position sample database: the method comprises the steps of collecting sample video data in a gallery bridge working area in an airport through a video collection system, wherein the sample video data comprises gallery bridge starting sample video data, gallery bridge stopping sample video data, gallery bridge starting sample video data and gallery bridge working ending sample video data, the video collection system transmits the sample video data to the data collection system, a data image frame collection module of the data collection system collects sample video image frames one by one on the sample video data and transmits the sample video image frames to a data image frame electronic line dividing module, the data image frame electronic line dividing module divides a sample video image frame into an electronic line A and an electronic line B, the electron beam A divides the sample video image frame into an A1 region and an A2 region, the electron beam B divides the A2 region of the sample video image frame into an A2B1 region and an A2B2 region, and the A2B1 region is close to the A1 region; the convolutional neural network system extracts and identifies sample video image frames to obtain a gallery bridge characteristic region and a background characteristic region, the sample video image frames are subjected to binarization processing and feature unification and then transmitted to a gallery bridge working state judgment module, and the gallery bridge working state judgment module obtains the working state of a gallery bridge through image feature extraction, comparison processing and logic judgment and transmits and stores the working state in a gallery bridge position sample database; the method for judging the gallery bridge working state of the sample video image frame by the gallery bridge working state judging module comprises the following steps:
a1, if the center line of the gallery bridge feature region of the continuous M-second sample video image frames is in the A1 region, judging that the gallery bridge does not work;
a2, if the center line of the gallery bridge feature region of the sample video image frame enters the A2B1 region from the A1 region and the center line of the gallery bridge feature region of the sample video image frame for N seconds continuously stays in the A2B1 region, judging that the working state of the gallery bridge is that the gallery bridge starts to approach the bridge, and correspondingly storing the sample video image frame at the position as gallery bridge starting to approach the bridge sample video data of a gallery bridge position sample database;
a3, if the center line of the gallery bridge feature region of the sample video image frame enters the A2B2 region from the A2B1 region and the center line of the gallery bridge feature region of the sample video image frame for N seconds continuously stays in the A2B2 region, judging that the working state of the gallery bridge is that the gallery bridge is closed, and then correspondingly storing the sample video image frame at the position as gallery bridge closing bridge finishing sample video data of a gallery bridge position sample database;
a4, if the center line of the gallery bridge feature region of the sample video image frame enters the A2B1 region from the A2B2 region and the center line of the gallery bridge feature region of the sample video image frame for N seconds continuously stays in the A2B1 region, judging that the gallery bridge starts to leave in the working state, and correspondingly storing the sample video image frame at the working state as the gallery bridge of the gallery bridge position sample database and starting to leave the sample video data;
a5, if the center line of the gallery bridge feature region of the sample video image frame enters the A1 region from the A2B1 region and the center line of the gallery bridge feature region of the sample video image frame for N seconds continuously stays in the A1 region, judging that the working state of the gallery bridge is the gallery bridge working end, and correspondingly storing the sample video image frame at the position as gallery bridge working end sample video data of a gallery bridge position sample database;
B. the method comprises the steps that real-time video data in a working area of a gallery bridge in an airport are collected through a video collection system, the video collection system transmits the real-time video data to the data collection system, a data image frame collection module of the data collection system collects real-time video image frames one by one and transmits the real-time video image frames to a data image frame electronic line dividing module, the data image frame electronic line dividing module divides the real-time video image frames into an electronic line A and an electronic line B, the electronic line A divides the real-time video image frames into an A1 area and an A2 area, the electronic line B divides an A2 area of the real-time video image frames into an A2B1 area and an A2B2 area, and the A2B1 area is close to an A1 area; the convolutional neural network system extracts and identifies real-time video image frames to obtain a gallery bridge characteristic region and a background characteristic region, and transmits the real-time video image frames to a gallery bridge working state judgment module after binarization processing, image characteristic extraction and characteristic unification;
C. the corridor bridge working state data are obtained, and the corridor bridge working state judgment module is used for obtaining the corridor bridge working state through image feature extraction, comparison processing and logic judgment according to the real-time video data, wherein the corridor bridge working state comprises the corridor bridge starting to approach the corridor bridge, the corridor bridge approach the corridor bridge finishing, the corridor bridge starting to leave and the corridor bridge finishing; the corridor bridge working state judgment method of the corridor bridge working state judgment module comprises the following steps:
c1, if the center line of the gallery bridge feature region of the continuous M-second real-time video image frames is in the A1 region, judging that the gallery bridge does not work;
c2, if the center line of the gallery bridge feature region of the real-time video image frame enters the A2B1 region from the A1 region and the center line of the gallery bridge feature region of the continuous N-second real-time video image frame stays in the A2B1 region, judging that the operating state of the gallery bridge is that the gallery bridge starts to lean against the bridge;
c3, if the center line of the gallery bridge feature region of the real-time video image frame enters the A2B2 region from the A2B1 region and the center line of the gallery bridge feature region of the continuous N-second real-time video image frame stays in the A2B2 region, judging that the working state of the gallery bridge is that the gallery bridge is closed;
c4, if the center line of the gallery bridge feature region of the real-time video image frame enters the A2B1 region from the A2B2 region and the center line of the gallery bridge feature region of the continuous N-second real-time video image frame stays in the A2B1 region, judging that the operating state of the gallery bridge is that the gallery bridge starts to leave;
and C5, if the center line of the gallery bridge feature region of the real-time video image frame enters the A1 region from the A2B1 region and the center line of the gallery bridge feature region of the continuous N-second real-time video image frame stays in the A1 region, judging that the working state of the gallery bridge is the end of the gallery bridge.
5. The gallery bridge working state detection method based on the image processing technology as claimed in claim 3 or 4, wherein: the sample video data in the step A comprises gallery and bridge position sample data under the weather conditions of night, rain, snow, fog or haze and sand dust, wherein the gallery and bridge position sample data under the weather conditions of night is not less than 10%, the gallery and bridge position sample data under the weather conditions of rain and snow is not less than 10%, the gallery and bridge position sample data under the weather conditions of fog or haze is not less than 10%, and the gallery and bridge position sample data under the weather conditions of sand dust is not less than 10%.
6. The gallery bridge working state detection method based on the image processing technology as claimed in claim 3 or 4, wherein: the video acquisition system in the step A and the step B comprises a plurality of video cameras, and all the video cameras are respectively arranged at the positions right in front, left in front, right above, left above and back above the corridor bridge working area in the airport; the data acquisition system in the step A and the data acquisition system in the step B acquire single video data of all the video cameras, combine the single video data into video data and transmit the video data to the data image frame electronic line dividing module; the video camera is a digital camera, and the resolution of the digital camera is not lower than 1080 multiplied by 720.
7. The gallery bridge working state detection method based on the image processing technology as claimed in claim 3 or 4, wherein: the number of the pictures of the sample video image frames of the gallery bridge position sample database is not less than ten thousand.
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 CN110210427A (en) | 2019-09-06 |
| CN110210427B true 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) |
Families Citing this family (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 |
| CN112528729B (en) * | 2020-10-19 | 2024-09-27 | 浙江大华技术股份有限公司 | Video-based aircraft bridge 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 |
| CN115410418B (en) * | 2022-10-31 | 2023-01-24 | 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 |
Family Cites Families (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 |
| CN104966045B (en) * | 2015-04-02 | 2018-06-05 | 北京天睿空间科技有限公司 | Aircraft disengaging berth automatic testing method based on video |
| 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 |
| 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 |
| US10480939B2 (en) * | 2016-01-15 | 2019-11-19 | Fugro Roadware Inc. | High speed stereoscopic pavement surface scanning system and method |
| CN107437245B (en) * | 2017-06-26 | 2020-10-20 | 西南交通大学 | High-speed railway contact net fault diagnosis method based on deep convolutional neural network |
| 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 |
-
2019
- 2019-06-06 CN CN201910489381.7A patent/CN110210427B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN110210427A (en) | 2019-09-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110210427B (en) | Corridor bridge working state detection system and method based on image processing technology | |
| CN109878552B (en) | Foreign matter monitoring devices between track traffic platform door and train based on machine vision | |
| CN101510356B (en) | Video detection system and data processing device thereof, video detection method | |
| CN102665071B (en) | Intelligent processing and search method for social security video monitoring images | |
| KR101793636B1 (en) | Image processing system | |
| CN104157138A (en) | Urban road video monitoring system | |
| TW202013252A (en) | License plate recognition system and license plate recognition method | |
| CN107808392B (en) | Automatic tracking and positioning method and system for security check vehicle in open scene | |
| CN102509090B (en) | A kind of vehicle feature recognition device based on public safety video image in sky net engineering | |
| CN109711318B (en) | Multi-face detection and tracking method based on video stream | |
| WO2021139049A1 (en) | Detection method, detection apparatus, monitoring device, and computer readable storage medium | |
| CN113947731B (en) | Foreign matter identification method and system based on contact net safety inspection | |
| Kulkarni et al. | Real time vehicle detection, tracking and counting using Raspberry-Pi | |
| CN103236181A (en) | Traffic signal lamp state monitoring system and method based on machine vision | |
| CN113011252B (en) | Rail foreign matter intrusion detection system and method | |
| CN104363426A (en) | Traffic video monitoring system and method with target associated in multiple cameras | |
| JP2002357557A (en) | Control system of road, surrounding facility, and the like | |
| CN111738203A (en) | Platform pedestrian line crossing detection method based on automatic line marking area | |
| CN113788051A (en) | Train on-station running state monitoring and analyzing system | |
| CN112434827A (en) | Safety protection identification unit in 5T fortune dimension | |
| CN203812397U (en) | Urban road video monitor system | |
| KR102169211B1 (en) | apparatus and method for automatically detecting bird's cast | |
| CN115909223A (en) | Method and system for matching WIM system information with monitoring video data | |
| CN110211159A (en) | A kind of aircraft position detection system and method based on image/video processing technique | |
| CN114359860A (en) | Track identification method and system based on environment perception instance segmentation model |
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 |