CN111210457B - Aircraft listing method combining video analysis and positioning information - Google Patents

Aircraft listing method combining video analysis and positioning information Download PDF

Info

Publication number
CN111210457B
CN111210457B CN202010017019.2A CN202010017019A CN111210457B CN 111210457 B CN111210457 B CN 111210457B CN 202010017019 A CN202010017019 A CN 202010017019A CN 111210457 B CN111210457 B CN 111210457B
Authority
CN
China
Prior art keywords
target
positioning
matching
analysis
information
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
Application number
CN202010017019.2A
Other languages
Chinese (zh)
Other versions
CN111210457A (en
Inventor
吴刚
林姝含
郑文涛
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Tianrui Kongjian Technology Co ltd
Original Assignee
Beijing Tianrui Kongjian Technology Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Beijing Tianrui Kongjian Technology Co ltd filed Critical Beijing Tianrui Kongjian Technology Co ltd
Priority to CN202010017019.2A priority Critical patent/CN111210457B/en
Publication of CN111210457A publication Critical patent/CN111210457A/en
Application granted granted Critical
Publication of CN111210457B publication Critical patent/CN111210457B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/20—Analysis of motion
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/70—Determining position or orientation of objects or cameras
    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10016—Video; Image sequence
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V2201/00—Indexing scheme relating to image or video recognition or understanding
    • G06V2201/07—Target detection

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Radar Systems Or Details Thereof (AREA)
  • Traffic Control Systems (AREA)

Abstract

The invention relates to an aircraft listing method combining video analysis and positioning information, which comprises the steps of carrying out target detection and tracking on a panoramic video image to obtain all analysis targets in the video image; matching the analysis target with the positioning target, introducing historical matching information into the matching cost function, listing according to the matching result, and updating the positioning target list according to the matching result. The invention is mainly used for the aircraft listing at the airport, can effectively avoid or reduce the phenomenon of 'listing', and can better realize the correct and continuous listing effect under the condition that the updating of the positioning information is slow, even the positioning system is closed.

Description

Aircraft listing method combining video analysis and positioning information
Technical Field
The invention relates to an aircraft listing method combining video analysis and positioning information, and belongs to the technical field of computer vision.
Background
The panoramic video splicing technology is widely applied to airport commanding, images shot by a plurality of cameras and having overlapping areas are spliced into high-resolution panoramic images in real time, the airport full-view is visually presented, and the requirement that a user can see the airport full-view is met. But with panoramic video only, the airport controller does not know the specific information of the aircraft in the video. Clearly, if the flight information of an aircraft (airplane) can be displayed in a video, great convenience is brought to airport command work. For this purpose, the aircraft listing technology further solves the problem of "what" on the basis of "what is seen", and displays the identification information of the aircraft in a video image, namely, a listing.
The existing aircraft listing methods mainly comprise two types, one type is that the aircraft latitude and longitude coordinates in the positioning information are directly mapped to a video image coordinate system to serve as the position of the aircraft in a video monitoring image by using the positioning information, and the aircraft listing is carried out. Currently, a positioning information system commonly used for monitoring an aircraft is an Automatic Dependent Surveillance-Broadcast (ADS-B) system, which directly obtains information such as a flight number, longitude and latitude of the aircraft, and automatically obtains information such as a position, an altitude, a speed, a course, an identification number and the like of the aircraft from relevant airborne equipment without manual operation or inquiry, so as to monitor the state of the aircraft by an airport controller. The listing method completely depends on the accuracy and real-time performance of a positioning system, and because positioning information is sent at intervals, flight positions mapped to video images cannot be accurately superposed on corresponding aircraft targets in the video, and listing display jumping can occur.
Another type of tag for determining the real-time position of an aircraft in a video image by combining positioning information with video analysis[1]And detecting and tracking the aircraft target in the video, and matching and associating the aircraft target with the positioning information to realize listing. Due to the utilization of the video analysis result, the flight information can be accurately superposed to the aircraft target in the video and can smoothly move along with the target, and the jumping problem of the former method is solved.
However, despite the clear advantages of combining location information with video analysis over relying solely on location information, the following problems remain:
1) analyzing the problem of string cards when the target is close: since the matching between the video analysis target and the positioning information is based only on the position information, and there is an error in the positioning information itself and its mapping to the image, the aviation sign may not be hung on the correct target when there are multiple analysis targets approaching. For example: when an original target which is correctly marked approaches another target in the process of moving, the mapping coordinate of the positioning information of the target is possibly closer to the target and matched with the target, and the phenomenon of 'string of cards' occurs; when multiple objects are close together, the sign may also randomly jump between the objects.
2) The problem of listing when the update of the positioning information is slow or missing: the update interval of the positioning information is preferably within 3 seconds, but may be long in practical application, sometimes reaching tens of seconds or even tens of seconds. For an analysis target which moves rapidly in a video, the analysis target is easy to cause that the analysis target cannot be matched correctly due to the fact that the analysis target is too far away from positioning information; in addition, the positioning information may be turned off after the aircraft is in a landing and taxiing state for a period of time, resulting in the failure to continue to hang tags.
Disclosure of Invention
The invention provides an aircraft listing method combining video analysis and positioning information, which overcomes the defects in the prior art, avoids or reduces the phenomenon of listing when multiple targets are close to each other, and can still carry out more effective listing when the positioning information cannot be obtained in time.
The technical scheme of the invention is as follows: an aircraft listing method combining video analysis and positioning information comprises the steps of carrying out target detection and tracking on a panoramic video image to obtain all targets in the video image, wherein the targets obtained through the video image are called analysis targets, each analysis target is endowed with an id (identification code), and the id of the same target is kept unchanged; acquiring positioning information, wherein a target acquired through the positioning information is called a positioning target, the positioning information comprises a flight number and longitude and latitude coordinates of the positioning target, and the longitude and latitude coordinates of the positioning target are mapped into an image coordinate system to acquire an image coordinate of the positioning target; matching an analysis target in a current frame with a current effective positioning target, and defining a matching cost function cost (i, j) of an analysis target i and a positioning target j according to the following formula:
cost(i,j)＝dist(i,j)*k(i,j)
wherein the content of the first and second substances,
dist (i, j) is the image distance between the analysis target i and the positioning target j, and the calculation formula is as follows:
Figure GDA0002943745930000031
k (i, j) is an attenuation coefficient, and the calculation formula is as follows:
k(i,j)＝αmatch_time(i,j)；
wherein the content of the first and second substances,
Figure GDA0002943745930000032
image coordinates of a central point of an analysis target i;
Figure GDA0002943745930000033
mapping the position data of the positioning target j to image coordinates on the image;
h is the height of the image;
match _ time (i, j) is the number of times that the analysis target i and the positioning target j have been continuously matched before the current matching, and if the analysis target i and the positioning target j have not been continuously matched before the current matching, the analysis target i and the positioning target j are not continuously matched
match_time(i,j)＝0；
Alpha is a set coefficient, and a positive number smaller than 1 is taken;
i is 1, …, and N is the number of analysis targets;
i is 1, …, M is the number of positioning targets,
taking the matching mode with the maximum matching target and the minimum sum of matching costs (a calculation value of a matching cost function) as a matching result, and performing aircraft listing according to the following modes:
1) if the analysis target i is matched with the positioning target j, according to the position of the analysis target
Figure GDA0002943745930000034
Width wiHigh h, hiAnd flight number flight of the positioning targetjHang cards, i.e. in order
Figure GDA0002943745930000041
Is an output result;
2) if the analysis target i does not have a matching positioning target, based on the position of the analysis target
Figure GDA0002943745930000042
Width wiHigh h, hiAnd id (id)i) Hang cards, i.e. in order
Figure GDA0002943745930000043
Is an output result;
3) if the positioning target j has no matching analysis target, the position of the positioning target is used
Figure GDA0002943745930000044
Default width wdefaultHigh by default hdefaultAnd flight number flightjHang cards, i.e. in order
Figure GDA0002943745930000045
Is the output result.
The matching operation can be performed generally using the Kuhn-Munkres algorithm.
Preferably, a matching cost threshold is set, and it is defined that the matching cost of any pair of analysis target and positioning target matching is not greater than the matching cost threshold.
And matching the analysis target with the positioning target for each frame of image, and updating the listing according to the matching result. Or, the matching of the analysis target and the positioning target is carried out once every other frame or a plurality of frames, and the listing is updated according to the matching result.
Preferably, a positioning target list is established and maintained, and the current positioning information is recorded.
The entries of the positioning list include the flight number and the longitude and latitude coordinates of the positioning target.
When a new piece of positioning information is obtained, retrieving a positioning target list through a flight number, and if the flight number exists, updating the corresponding longitude and latitude coordinates; if the flight number does not exist, adding an entry, and recording the flight number and the longitude and latitude coordinates of the positioning target; if a certain table entry is not updated after a certain time, deleting the table entry, and when the analysis target is matched with the positioning target, taking all positioning information recorded by the current positioning target list as the positioning information of the positioning target and obtaining the positioning information from the positioning target list.
Updating the positioning target list according to the matching result, and if the analysis target i is successfully matched with the positioning target j, matching the analysis target i with the positioning target j
Figure GDA0002943745930000046
Conversion to latitude and longitude coordinates
Figure GDA0002943745930000047
If the flight number already stored in the positioning target list is flightjTo table entry of
Figure GDA0002943745930000051
Updating the longitude and latitude coordinates in the table item, otherwise adding a new table item, and recording the flight number flightjAnd latitude and longitude coordinates
Figure GDA0002943745930000052
And broadcasting the information of the positioning target list, wherein the positioning target list is broadcasted once every updating or every certain time.
The invention has the beneficial effects that: due to the introduction of the historical matching information, the matching cost function is optimized, and the phenomenon of 'plate string' when the target is close is effectively avoided or reduced; due to the fact that the updating mode of the positioning information is optimized, the listing can be well achieved even under the condition that the positioning information is updated slowly and the positioning system is closed, and the correct and continuous listing effect is well achieved.
Drawings
FIG. 1 is a schematic overall flow diagram of the present invention;
FIG. 2 is a schematic diagram of a listing matching process of the present invention;
fig. 3 is a screenshot of aircraft listing results.
Detailed Description
The following describes the overall flow of the aircraft listing method combining video analysis and positioning information in a summary manner, and then describes the improvement of the key listing matching method of the present invention in detail.
1. Integrated process
The overall flow of the process of the invention is shown in FIG. 1.
The input is panoramic video and positioning information. The panoramic video is obtained through a video splicing technology and is used for video target analysis; the positioning information can be from an ADS-B system or other positioning information systems (or called positioning systems), at least comprises longitude and latitude information and flight information (flight number or flight identification information in other forms) of the aircraft, and the longitude and latitude information and the flight information are matched and associated to obtain the flight information of each aircraft target in the video, so that flight listing is realized.
For the panoramic video image, an aircraft target in the panoramic video image can be obtained through real-time analysis by adopting a target detection and tracking technology, and an analysis result is represented by a circumscribed rectangular frame (analysis frame) of the target and contains position and size information of the target; in order to distinguish different targets, each analysis frame is provided with an id, and the id of the same target can be kept unchanged in the motion process by utilizing a multi-target tracking technology. The current mainstream technology for detecting and tracking targets in video images, such as YOLO, can be adopted[2]、SORT[3]And the like.
For the positioning information, the longitude and latitude are mapped into image coordinates so as to be matched with the video analysis result. And judging whether each piece of positioning information received from the positioning system falls within the latitude and longitude range of the current panoramic image, if so, carrying out coordinate mapping, and otherwise, ignoring the information. The panoramic latitude and longitude range and the coordinate mapping need to be calibrated in advance. The coordinate mapping needs to select a plurality of reference points in the panoramic image and record the image coordinates of the reference points and the longitude and latitude of the corresponding position. When receiving new positioning information, the image coordinate of the positioning target can be estimated by a least square method by utilizing the longitude and latitude (or called longitude and latitude coordinates) of the known reference point[1]。
And then, performing matching association according to the position relationship between the analysis result and the positioning information, displaying the flight information beside the corresponding aircraft in the video image according to a certain rule, so as to realize the tag listing of the aircraft, wherein the tag listing effect is shown in fig. 3, and the word string beside the target box represents the flight number. If other information (such as the model of the airplane, the movement speed, etc.) is also included in the positioning information, the positioning information can be displayed in the result.
2. Matching method
The overall process of listing matching is shown in fig. 2. The matching is performed for each frame of image, and the input comprises:
(1) the ith aircraft analysis target of the current image frame, denoted as
Figure GDA0002943745930000061
Wherein
Figure GDA0002943745930000062
wi,hiAnd idiAnd respectively representing the coordinate, width, height and id of the central point image of the analysis target, wherein N is the number of targets obtained by analyzing the current frame.
On the time axis, objects with the same id represent the same object.
(2) The positioning target at the current moment is recorded as
Figure GDA0002943745930000063
Wherein
Figure GDA0002943745930000064
flightjAnd representing the image coordinates and flight numbers after the longitude and latitude coordinates in the positioning information are mapped. In practical applications, the update frequency of the positioning information (updated once in 3-5 seconds or even longer) is generally much lower than the video frame rate (25 fps).
The system (system for implementing the method of the invention) maintains a positioning target list, records the current positioning information, when a new piece of positioning information (including flight number and longitude and latitude) is obtained, the list is searched through the flight number, and if the flight number exists, the corresponding longitude and latitude information is updated; if not, adding a new table entry; if a certain table entry is not updated after a certain time, deleting the table entry. And when the cards are matched, for each frame of image, acquiring all the table entries in the positioning target list, and after the longitude and latitude mapping is completed, starting to be matched with the analysis target.
The listing matching can be abstracted into a bipartite graph optimal weight matching problem. The problem can be solved by the Kuhn-Munkres algorithm in graph theory[4]To solve the problem. Given N analysis targets and M localization targets and their matching cost functions cost (i, j), i 1, …, N, j 1, …, M, the best match is found, with the most matching targets and the least sum of matching costs.
For the matching results obtained by the Kuhn-Munkres algorithm, further filtering is needed, and when the matching cost between the matching targets exceeds a certain threshold, the matching relationship between the matching targets should be released so as to filter out the target pairs with excessive matching cost.
The final matching result includes the following three cases:
1) the analysis target i is successfully matched with the positioning target j by using the position, the width and the height of the analysis target and the flight number of the positioning target, namely
Figure GDA0002943745930000071
As an output result.
2) Location target with which analysis target i does not match, using location, width and height of analysis target i
Figure GDA0002943745930000072
As an output result.
3) Analysis targets with which the location target j does not match, using location, default width and flight number of the location target, i.e. flight number
Figure GDA0002943745930000073
As an output result.Because the target is not located with target width and height information, a predefined default value is used
wdefault,hdefaultAnd (4) showing.
The following further describes the definition of the matching cost function and the updating method of the positioning information.
(1) Definition of matching cost function
Intuitively, the closer the analysis target and the positioning target are, the more likely the analysis target and the positioning target are to be matched, but as mentioned above, because there is inevitably a certain error in the positioning information itself (latitude and longitude) and its mapping process, when a certain analysis target approaches to other targets in the moving process, or when a plurality of analysis targets are closer in distance, the matching may not be stable enough, which may cause the flight tag to move between different targets. In order to make the listing more stable, historical listing information is introduced into the matching cost function:
cost (i, j) ═ dist (i, j) × k (i, j) formula (1)
Wherein dist (i, j) represents the distance between the analysis target i and the positioning target j, and k (i, j) is an attenuation coefficient determined by the history matching information of i and j, and can be specifically defined as follows:
Figure GDA0002943745930000081
the former term in the formula is the Euclidean distance between the analysis target and the positioning target; the latter term is a perspective coefficient, and due to the perspective relationship of the near term and the far term, the distances in the real scene represented by the same distance at different image positions are different, and are usually larger above the image (the y coordinate is small); and generally smaller below the image (large y-coordinate). Therefore, the ratio of 1 minus the y coordinate of the midpoint of the target link to the image height H is used as the perspective coefficient. k (i, j) represents the matching cost attenuation coefficient between the analysis target i and the positioning target j, and can be defined as follows:
k(i,j)＝αmatch_time(i,j)formula (3)
Wherein, match _ time (i, j) represents the number of times that the analysis object i and the positioning object j have been continuously matched, α is a positive number (e.g. 0.95) smaller than 1, k (i, j) decreases with the increase of the number of continuous matching times between i and j; if i and j are not matched continuously, and match _ time (i, j) is 0, k (i, j) is 1, and the matching cost is not attenuated. In this way, the match relationship is advantageously maintained between already matched objects, thereby reducing the "stringing" phenomenon.
(2) Optimization of positioning information updating mode
Under the conditions that extra matching rules are not introduced excessively and the matching frames are kept simple and consistent, the problem of listing caused by slow updating or closing of the positioning information can be improved by optimizing the updating mode of the positioning information. As mentioned above, the positioning information is stored in the positioning target list, and each entry includes flight number and longitude and latitude information. If the positioning information is updated slowly, the deviation between the target and the current actual position is large, and the target and the current actual position are difficult to be matched correctly; if the list item is not updated for a long time, the list item is deleted, and the listing cannot be continuously carried out.
In order to update the positioning information in time, a data source is added to the positioning target list, and besides the real positioning system (such as ADS-B), the data source can also be from the history matching result. If the positioning target j is successfully matched with the analysis target i, matching information is used
Figure GDA0002943745930000091
Updating the positioning information list, if the flight number in the list is flightjUpdating the position information of the table entry; otherwise, adding new table item. Before updating, the image coordinates need to be updated
Figure GDA0002943745930000092
And performing inverse mapping and converting into corresponding longitude and latitude coordinates. The reflection mapping method is completely similar to the positive mapping method, and interpolation calculation is carried out by using image coordinates of a plurality of pre-calibrated reference points and longitude and latitude coordinates corresponding to the image coordinates. Therefore, when the next frame is matched, the positioning information can be changed along with the history matching result even if the positioning information is not updated by the real positioning system, and the larger deviation between the positioning information and the real position of the target can be avoided. When the positioning system is turned off, the target can be continuously detected and trackedAnd after the history matching result updates the positioning information list, continuous listing can be realized.
The fact that the image coordinates of the matching target are not adopted to directly update the positioning information list, the inverse mapping is carried out firstly, and the fact that longitude and latitude coordinates are absolute coordinates with uniqueness is considered. When a plurality of panoramic videos with certain overlapped areas are analyzed and listed, if the reversely calculated longitude and latitude information is broadcasted, each panoramic can be received, and as long as the panoramic falls within the longitude and latitude range of the current panoramic, the coordinate mapping and matching can be carried out. In this way, the cross-panoramic continuous listing of the moving target can be realized.
The technical means disclosed by the invention can be combined arbitrarily to form a plurality of different technical schemes except for special description and the further limitation that one technical means is another technical means.
Reference to the literature
[1] Thangong et al, "a new method for realizing automatic tag hanging of airplane in airport video", Jiangsu university newspaper (Nature science edition), Vol.34, No. 6, 2013.
[2]Joseph Redmon and Ali Farhadi,“YOLOv3:An Incremental Improvement”,Technical report,2018.
[3]Alex Bewley,et al.,“Simple Online and Real-time Tracking”,International Conference on Image Processing(ICIP),2016.
[4] High body of detail, "graph theory and network flow theory," advanced education press, 2009.

Claims (6)

1. An aircraft listing method combining video analysis and positioning information comprises the steps of carrying out target detection and tracking on a panoramic video image to obtain all targets in the video image, wherein the targets obtained through the video image are called analysis targets, each analysis target is endowed with an id, and the ids of the same target are kept unchanged; acquiring positioning information, wherein a target acquired through the positioning information is called a positioning target, the positioning information comprises a flight number and longitude and latitude coordinates of the positioning target, and the longitude and latitude coordinates of the positioning target are mapped into an image coordinate system to acquire an image coordinate of the positioning target; matching an analysis target in a current frame with a current effective positioning target, and defining a matching cost function cost (i, j) of an analysis target i and a positioning target j according to the following formula:
cost(i，j)＝dist(i，j)*k(i，j)
wherein the content of the first and second substances,
dist (i, j) is the image distance between the analysis target i and the positioning target j, and the calculation formula is as follows:
Figure FDA0002943745920000011
k (i, j) is an attenuation coefficient, and the calculation formula is as follows:
k(i，j)＝αmatch_time(i，j)；
wherein the content of the first and second substances,
Figure FDA0002943745920000012
image coordinates of a central point of an analysis target i;
Figure FDA0002943745920000013
mapping the positioning data of the positioning target j to image coordinates on the image;
h is the height of the image;
the match _ time (i, j) is the number of times that the analysis target i and the positioning target j are continuously matched before the current matching, and if the analysis target i and the positioning target j are not continuously matched before the current matching, the match _ time (i, j) is 0;
alpha is a set coefficient, and a positive number smaller than 1 is taken;
i 1, N is the number of analysis targets;
j is 1, M is the number of positioning targets,
taking the matching mode with the maximum matching target and the minimum sum of the matching costs as a matching result, and carrying out aircraft listing according to the following modes:
1) if it is for the purpose of analysisThe mark i is matched with a positioning target j, and the position of the target is analyzed according to the mark i
Figure FDA0002943745920000021
Width wiHigh h, hiAnd flight number flight of the positioning targetjHang cards, i.e. in order
Figure FDA0002943745920000022
Is an output result;
2) if the analysis target i does not have a matching positioning target, based on the position of the analysis target
Figure FDA0002943745920000023
Width wiHigh h, hiAnd id listing, i.e. by
Figure FDA0002943745920000024
Is an output result;
3) if the positioning target j has no matching analysis target, the position of the positioning target is used
Figure FDA0002943745920000025
Default width wdefaultHigh by default hdefaultAnd flight number flightjHang cards, i.e. in order
Figure FDA0002943745920000026
Is the output result.
2. The method of claim 1, wherein a Kuhn-Munkres algorithm is adopted to perform matching operation, and a matching cost threshold is set, so as to limit the matching cost of any pair of the analysis target and the positioning target not to be greater than the matching cost threshold.
3. The method according to claim 2, wherein the matching of the analysis target with the positioning target is performed for each frame of image and the listing is updated according to the matching result, or the matching of the analysis target with the positioning target is performed once every other frame or frames and the listing is updated according to the matching result.
4. A method according to any one of claims 1-3, characterized by establishing and maintaining a list of positioning objects, recording current positioning information, the entries of the positioning list comprising the flight number and the longitude and latitude coordinates of the positioning object, retrieving the list of positioning objects by the flight number when a new piece of positioning information is obtained, and updating the corresponding longitude and latitude coordinates if the flight number exists; if the flight number does not exist, adding an entry, and recording the flight number and the longitude and latitude coordinates of the positioning target; if a certain table entry is not updated after a certain time, deleting the table entry, and when the analysis target is matched with the positioning target, taking all positioning information recorded by the current positioning target list as the positioning information of the positioning target and obtaining the positioning information from the positioning target list.
5. The method of claim 4, wherein the list of positioning targets is updated with the matching results, and if the matching between the analysis target i and the positioning target j is successful, the matching results are obtained
Figure FDA0002943745920000031
Conversion to latitude and longitude coordinates
Figure FDA0002943745920000032
If the flight number already stored in the positioning target list is flightjTo table entry of
Figure FDA0002943745920000033
Updating the longitude and latitude coordinates in the table item, otherwise adding a new table item, and recording the flight number flightjAnd latitude and longitude coordinates
Figure FDA0002943745920000034
6. The method of claim 5, wherein the location target list information is broadcast, wherein the location target list is broadcast once every update, or once every certain time.
CN202010017019.2A 2020-01-08 2020-01-08 Aircraft listing method combining video analysis and positioning information Active CN111210457B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010017019.2A CN111210457B (en) 2020-01-08 2020-01-08 Aircraft listing method combining video analysis and positioning information

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010017019.2A CN111210457B (en) 2020-01-08 2020-01-08 Aircraft listing method combining video analysis and positioning information

Publications (2)

Publication Number Publication Date
CN111210457A CN111210457A (en) 2020-05-29
CN111210457B true CN111210457B (en) 2021-04-13

Family

ID=70788994

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010017019.2A Active CN111210457B (en) 2020-01-08 2020-01-08 Aircraft listing method combining video analysis and positioning information

Country Status (1)

Country Link
CN (1) CN111210457B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113763416A (en) * 2020-06-02 2021-12-07 璞洛泰珂(上海)智能科技有限公司 Automatic labeling and tracking method, device, equipment and medium based on target detection
CN113313733A (en) * 2021-05-19 2021-08-27 西华大学 Hierarchical unmanned aerial vehicle target tracking method based on shared convolution

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130342380A1 (en) * 2005-04-20 2013-12-26 Accipiter Radar Technologies Inc. Low cost, high performance radar networks
CN104992451A (en) * 2015-06-25 2015-10-21 河海大学 Improved target tracking method
US20160275802A1 (en) * 2015-03-20 2016-09-22 Northrop Grumman Systems Corporation Unmanned aircraft detection and targeting of other aircraft for collision avoidance
CN108133028A (en) * 2017-12-28 2018-06-08 北京天睿空间科技股份有限公司 It is listed method based on the aircraft that video analysis is combined with location information
CN109615937A (en) * 2019-01-18 2019-04-12 南京航空航天大学 Segment runing time optimizes computing device
CN109635657A (en) * 2018-11-12 2019-04-16 平安科技（深圳）有限公司 Method for tracking target, device, equipment and storage medium
CN110163034A (en) * 2018-02-27 2019-08-23 冷霜 A kind of listed method of aircraft surface positioning extracted based on optimal characteristics

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104243935B (en) * 2014-10-10 2018-02-16 南京莱斯信息技术股份有限公司 Airport field prison aims of systems monitoring method based on video identification
CN105391975A (en) * 2015-11-02 2016-03-09 中国民用航空总局第二研究所 Video processing method in scene monitoring, device and scene monitoring system
CN108446634B (en) * 2018-03-20 2020-06-09 北京天睿空间科技股份有限公司 Aircraft continuous tracking method based on combination of video analysis and positioning information

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130342380A1 (en) * 2005-04-20 2013-12-26 Accipiter Radar Technologies Inc. Low cost, high performance radar networks
US20160275802A1 (en) * 2015-03-20 2016-09-22 Northrop Grumman Systems Corporation Unmanned aircraft detection and targeting of other aircraft for collision avoidance
CN104992451A (en) * 2015-06-25 2015-10-21 河海大学 Improved target tracking method
CN108133028A (en) * 2017-12-28 2018-06-08 北京天睿空间科技股份有限公司 It is listed method based on the aircraft that video analysis is combined with location information
CN110163034A (en) * 2018-02-27 2019-08-23 冷霜 A kind of listed method of aircraft surface positioning extracted based on optimal characteristics
CN109635657A (en) * 2018-11-12 2019-04-16 平安科技（深圳）有限公司 Method for tracking target, device, equipment and storage medium
CN109615937A (en) * 2019-01-18 2019-04-12 南京航空航天大学 Segment runing time optimizes computing device

Non-Patent Citations (6)

* Cited by examiner, † Cited by third party
Title
SIMPLE ONLINE AND REALTIME TRACKING;Alex Bewley等;《arXiv:1602.00763v2 [cs.CV]》;20170707;第1-5页 *
Stochastic air freight hub location and flight routes planning;Ta-Hui Yang;《Applied Mathematical Modelling》;20090321;第4424-4430页 *
YOLOv3: An Incremental Improvement;Joseph Redmon等;《arXiv:1804.02767v1 [cs.CV]》;20180408;第1-6页 *
一种在机场视频中实现飞机自动挂标牌的新方法;唐勇 等;《江苏大学学报（自然科学版）》;20131130;第34卷(第6期);第681-686页 *
基于图像识别的航空器场面运行监视技术研究;焦阳;《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》;20170315(第3期);C031-1792 *
基于统计方法的运动目标检测与跟踪技术研究;王勇;《中国博士学位论文全文数据库 信息科技辑》;20091115(第11期);I138-61 *

Also Published As

Publication number Publication date
CN111210457A (en) 2020-05-29

Similar Documents

Publication Publication Date Title
TWI795667B (en) A target tracking method, device, system, and computer accessible storage medium
US9363489B2 (en) Video analytics configuration
CN112488073A (en) Target detection method, system, device and storage medium
CN111210457B (en) Aircraft listing method combining video analysis and positioning information
US11657373B2 (en) System and method for identifying structural asset features and damage
US20140211987A1 (en) Summarizing salient events in unmanned aerial videos
CN101840422A (en) Intelligent video retrieval system and method based on target characteristic and alarm behavior
WO2022227764A1 (en) Event detection method and apparatus, electronic device, and readable storage medium
CN110290346B (en) Bidding video acquisition method based on intelligent video analysis
US11734338B2 (en) Image search in walkthrough videos
CN110703760B (en) Newly-added suspicious object detection method for security inspection robot
CN113177968A (en) Target tracking method and device, electronic equipment and storage medium
CN109636835B (en) Foreground object detection method based on template optical flow
CN109239702B (en) Airport low-altitude flying bird number statistical method based on target state set
CN110619308A (en) Aisle sundry detection method, device, system and equipment
CN115719436A (en) Model training method, target detection method, device, equipment and storage medium
CN113076899A (en) High-voltage transmission line foreign matter detection method based on target tracking algorithm
CN102291568A (en) Accelerated processing method of large-view-field intelligent video monitoring system
CN110688873A (en) Multi-target tracking method and face recognition method
CN109684953B (en) Method and device for pig tracking based on target detection and particle filter algorithm
CN114926859A (en) Pedestrian multi-target tracking method in dense scene combined with head tracking
CN114170556A (en) Target track tracking method and device, storage medium and electronic equipment
KR20230013297A (en) Methods and devices for automatic identification of changes in digital twin and image-based facilities
Wu et al. NDMFCS: An automatic fruit counting system in modern apple orchard using abatement of abnormal fruit detection
CN113312951A (en) Dynamic video target tracking system, related method, device and equipment

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