CN103488801B - A kind of airport target detection method based on geographical information space database - Google Patents
A kind of airport target detection method based on geographical information space database Download PDFInfo
- Publication number
- CN103488801B CN103488801B CN201310485693.3A CN201310485693A CN103488801B CN 103488801 B CN103488801 B CN 103488801B CN 201310485693 A CN201310485693 A CN 201310485693A CN 103488801 B CN103488801 B CN 103488801B
- Authority
- CN
- China
- Prior art keywords
- airport
- remote sensing
- characteristic curve
- aerial vehicle
- unmanned aerial
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/29—Geographical information databases
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Remote Sensing (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Image Processing (AREA)
- Processing Or Creating Images (AREA)
Abstract
The invention discloses a kind of airport target detection method based on geographical information space database, relate to target detection technique field in unmanned plane computer vision.This invention is for the test problems of airport target in unmanned aerial vehicle remote sensing images, by two, online and offline processing procedure, makes full use of remote sensing satellite and the prior information of geography information offer, it is achieved that airport target detects rapidly and accurately and positions.Under Xian, extract the attribute informations such as the direction of airport target, characteristic curve, characteristic curve distance according to satellite remote-sensing image and geography information；On line, to the remote sensing images obtained in real time according to the direction on doubtful airport current in geographical information space, with reference to the precision of image rectification in unmanned aerial vehicle remote sensing images processing system, extract the line feature in unmanned aerial vehicle remote sensing images, finally by characteristic curve on Comprehensive Evaluation unmanned aerial vehicle remote sensing images and the matching degree of doubtful airport target characteristic curve in geographical information space, thus the airport target detecting in real time and orienting in remote sensing images.
Description
Technical field
The present invention relates to target detection technique field in unmanned plane computer vision.Exactly,
The present invention relates to the priori that a kind of combining geographic information space provides, by unmanned aerial vehicle remote sensing shadow
The method realizing airport target detection as middle part extension set field areas.
Background technology
With unmanned plane as platform, by airborne target identification sensor with based on radio operation mesh
Mark not (RBCI) system, uses the collaborative technical system combined with miscoordination identification, it is possible to
Realizing battlefield to enter a war the target detection and identification of unit, target detection and identification is to support weapon system
System beyond-visual-range operation, an important means of auxiliary precision strike, be the information-based premise fought,
Also it is the premise of unmanned plane antagonism simultaneously.
Along with lifting and the development of image processing algorithm of hardware performance, both at home and abroad to unmanned aerial vehicle vision
The research of vision system has become the focus in unmanned plane field.Studies in China personnel mainly study mark
The detection of will thing is also used in Autonomous Landing of UAV and navigation aspect, and research uses target
Characteristic matching or statistical pattern recognition method.Such as geometry, color characteristic coupling.Hu is not
Bending moment, Haar feature classifiers etc..And external research lays particular emphasis on the scouting to Research on Target
With track and localization, groundwork herein is also the orientation problem studying this spot.At this
In class problem, due to the template characteristic that target is the most unified, so more difficult use is above-mentioned to mark
The method of analyte detection carrys out track and localization target, especially for this kind of large-scale target in airport,
Often can only observe the part on airport during unmanned plane during flying, tradition is based on line detection
Algorithm can lose efficacy, and for this, we have employed a kind of target detection based on geospatial information storehouse calmly
Position technology.
Summary of the invention
In view of the weak point in background technology, present invention is primarily targeted at by fully profit
With the satellite remote sensing resource become increasingly abundant at present and geographic information resources, it is possible to low at high-accuracy
The part airport being quickly detected from unmanned aerial vehicle remote sensing images, fall is realized under the requirement of false drop rate
The impact of low remote sensing image quality.A kind of airport based on geographical information space mesh is provided for this
Mark detection method, to solve in prior art that part airport target recall rate is low, false drop rate is high,
Poor real, the problem easily affected by remote sensing image quality.
For reaching above-mentioned purpose, the present invention provides a kind of airport based on geographical information space database mesh
Mark detection method, it is characterised in that: the method is divided into two, online and offline processing procedure；?
Under line, according to satellite remote-sensing image and geography information, build the geographical information space of airport target；
On line, the characteristic information of airport target in combining geographic information space, it is achieved unmanned aerial vehicle remote sensing
The detection of airport target and location in image, the method to realize step as follows:
Step S1: combine satellite remote-sensing image and the geographical position of airport target, sets up geography
Information space storehouse；
Step S2: according to positional information and the correction accuracy of unmanned aerial vehicle remote sensing images of unmanned plane
The airport attribute information corresponding with unmanned plane position is extracted, profit from geographical information space database
By the directional information on airport doubtful in geographical information space database, extracted by the method for line integral
Characteristic curve information on unmanned aerial vehicle remote sensing images；
Step S3: by the characteristic curve information on unmanned aerial vehicle remote sensing images and geographical information space database
In the attribute information of doubtful airport feature line carry out Comprehensive Evaluation, determine on unmanned aerial vehicle remote sensing images
The position of characteristic curve whether be the airport position corresponding with unmanned plane position；
Step S4: according to the attribute information of airport target in geographical information space database and characteristic curve
Positional information generate initial part airport profile, thus realize machine in unmanned aerial vehicle remote sensing images
The real-time detection of field target.
Wherein, the method setting up geographical information space database in step S1 is, first with man-machine
Mutual mode selects characteristic curve direction, then utilizes the method for line integral to extract airport target bag
The characteristic curve contained, finally utilizes characteristic curve attribute information to build geographical information space with geography information
Storehouse.
Wherein, step S2 is extracted on unmanned aerial vehicle remote sensing images by the method for line integral
Comprising the concrete steps that of characteristic curve:
Step 21: according to the geographical coordinate of unmanned aerial vehicle remote sensing images in geographical information space database
Search doubtful airport target；
Step 22: utilize first order differential operator to calculate unmanned aerial vehicle remote sensing images vertical gradient and water
Flat direction gradient；
Step 23: whether be perpendicular to airport side by the gradient judging unmanned aerial vehicle remote sensing images
Reach up to maximum to the doubtful point of characteristic curve judging on unmanned aerial vehicle remote sensing images；
Step 24: by the method for line integral, the characteristic curve on statistical machine field direction doubtful
Point number extracts the characteristic curve on unmanned aerial vehicle remote sensing images.
Wherein, whether the position determining the characteristic curve on unmanned aerial vehicle remote sensing images in step S3 is
The method of the airport position corresponding with unmanned plane position is: by global alignment unmanned plane
In remote sensing image, characteristic curve comes with the characteristic curve attribute information of airport target in geographical information space
Judge.
Wherein, the method generating initial part airport profile described in step S4 is: comprehensive
Utilize the characteristic curve extracted in the attribute of airport target in geographical information space and remote sensing image
Sketch out the accurate location on airport.
Beneficial effects of the present invention:
The present invention is different from other prior aries, and the present invention proposes one and utilizes satellite remote sensing to provide
The technology that airport target in unmanned aerial vehicle remote sensing images is detected by source and geographic information resources,
Part is solved by the priori making full use of remote sensing satellite and GIS-Geographic Information System offer
The problem that airport feature point is difficult to detect less.Compared with previous methods, the method that the present invention proposes
Accuracy further increases the recall rate of airport target in unmanned aerial vehicle remote sensing images, simultaneously
Reduce at the bottom of false drop rate and the computation complexity of algorithm, and be less susceptible to remote sensing image quality
Impact.
Accompanying drawing explanation
Fig. 1 is the workflow diagram of the embodiment of the present invention.
Detailed description of the invention
The workflow of the embodiment of the present invention is as it is shown in figure 1,1 skill of the present invention below in conjunction with the accompanying drawings
Each detailed problem involved in art scheme is further described.It is to be noted that described
Embodiment be intended merely to facilitate the understanding of the present invention and do not play any restriction effect.
A kind of airport target detection method based on geographical information space database, it is characterised in that: should
Method is divided into two, online and offline processing procedure；Under Xian, according to satellite remote-sensing image and ground
Reason information, builds the geographical information space of airport target；On line, combining geographic information space
The characteristic information of middle airport target, it is achieved in unmanned aerial vehicle remote sensing images, the detection of airport target is with fixed
Position, the method to realize step as follows:
Step S1: combine satellite remote-sensing image and the geographical position of airport target, sets up geography
Information space storehouse；
The method setting up geographical information space database in step S1 is, first with man-machine interaction
Mode selects characteristic curve direction, then utilizes the method for line integral to extract the spy that airport target comprises
Levy line, finally utilize characteristic curve attribute information to build geographical information space database with geography information.
Step S1 detailed description of the invention is as follows:,
Step 11: utilize edge detection operator, as Laplacian operator, Sobel operator,
Roberts operator, Prewitt operator etc. calculate satellite remote-sensing image horizontal and vertical two
Gradient information on direction.Especially, we use first-order difference method to enter gradient herein
Row calculates, and vertical gradient is designated asThe gradient of horizontal direction
Step 12: utilize the mode of man-machine interaction, selects To Airport as characteristic curve direction θ；
Step 13: according to the characteristic curve direction θ be given in step 12, it is judged that all pixels
Projection value on θ ± 90 ° is the most maximum, if maximum, is labeled as 1, is otherwise zero；
Step 14: along the direction θ that step 12 is given, utilizes integration method from different pixels
Statistic procedure of setting out 13 is labeled as the pixel number of 1, when statistical magnitude exceedes threshold value,
Then labelling current search line segment is characterized line segment；
Step 15: the characteristic curve obtained according to step 14, statistics the distribution of recording feature line
Situation, including the direction of characteristic curve, distance relation, characteristic line number etc.；
Step 16: combine the characteristic curve distribution feelings of record in the geographical coordinate on airport and step 15
Condition, builds geographical information space database.
Step S2: according to positional information and the correction accuracy of unmanned aerial vehicle remote sensing images of unmanned plane
The airport attribute information corresponding with unmanned plane position is extracted, profit from geographical information space database
By the directional information on airport doubtful in geographical information space database, extracted by the method for line integral
Characteristic curve information on unmanned aerial vehicle remote sensing images；
Wherein, step S2 is extracted on unmanned aerial vehicle remote sensing images by the method for line integral
Comprising the concrete steps that of characteristic curve:
Step 21: according to the geographical coordinate of unmanned aerial vehicle remote sensing images in geographical information space database
Search doubtful airport target；
Step 22: utilize first order differential operator to calculate unmanned aerial vehicle remote sensing images vertical gradient and water
Flat direction gradient；
Step 23: whether be perpendicular to airport side by the gradient judging unmanned aerial vehicle remote sensing images
Reach up to maximum to the doubtful point of characteristic curve judging on unmanned aerial vehicle remote sensing images；
Step 24: by the method for line integral, the characteristic curve on statistical machine field direction doubtful
Point number extracts the characteristic curve on unmanned aerial vehicle remote sensing images.
Step S3: by the characteristic curve information on unmanned aerial vehicle remote sensing images and geographical information space database
In the attribute information of doubtful airport feature line carry out Comprehensive Evaluation, determine on unmanned aerial vehicle remote sensing images
The position of characteristic curve whether be the airport position corresponding with unmanned plane position；
Wherein, whether the position determining the characteristic curve on unmanned aerial vehicle remote sensing images in step S3 is
The method of the airport position corresponding with unmanned plane position is: by global alignment unmanned plane
In remote sensing image, characteristic curve comes with the characteristic curve attribute information of airport target in geographical information space
Judge.
Step S3 mainly comprises step in detail below:
Step 31: near every the characteristic curve extracted in statistic procedure 24, characteristic line number is (special
The lowest, the bar number of all characteristic curves in our 10 pixels of statistical distance characteristic curve herein),
And possible characteristic line number contrasts with geographical information space database；
Step 32: the greatest length of the characteristic curve extracted in calculation procedure 24, and according to nothing
The resolution of man-machine remote sensing image, is converted to the distance of characteristic curve, if distance is more than 100 meters,
Then think that the doubtful airport in geographical information space database is implicitly present in；
Step 33: the centrage of the characteristic curve bunch extracted in extraction step 24, calculates feature
Linear distance, and compare with the distance of characteristic curve in geographical information space database, if characteristic curve
Standoff distance close, then it is assumed that there is airport；
Step 34: the airfield detection result that combining step 31,32 and 33 is given, it is judged that when
Whether front unmanned aerial vehicle remote sensing images exists doubtful airport.
Step S4: according to the attribute information of airport target in geographical information space database and characteristic curve
Positional information generate initial part airport profile, thus realize machine in unmanned aerial vehicle remote sensing images
The real-time detection of field target.
Wherein, the method generating initial part airport profile described in step S4 is: comprehensive
Utilize the characteristic curve extracted in the attribute of airport target in geographical information space and remote sensing image
Sketch out the accurate location on airport.
Step S4 comprises step in detail below:
Step 41: the fixed point of outermost end characteristic curve in all characteristic curves that selecting step 31 extracts,
It is sequentially connected the convex shape changeable sketched out as preliminary airport target；
Step 42: according to the average gray value of pixel on characteristic curve and variance, will tentatively extract
Traffic pattern extend along To Airport, until the pixel on gray value and the characteristic curve of pixel is poor
Different relatively big or till arriving image border (especially, this threshold value we be chosen as 3 times of features
The variance of grey scale pixel value on line)；
Step 43: the traffic pattern after extending in output step 42.
In sum, part airport target during the present invention proposes a kind of unmanned aerial vehicle remote sensing images
Detection method, satellite remote-sensing image and geography information as priori, are utilized line by this method
The strategy of integration extracts characteristic curve, uses the strategy of characteristic curve coupling to realize part airport target
Detect and position, it is possible to realize airport target in the situation that sound pollution is bigger the most accurate
Ground detection and location.
Although the preferred forms of the present invention illustrates the present invention, it being understood, however, that
On the premise of without departing substantially from claims defined value invention essence, the present invention can be done certain
A little amendments.
Claims (4)
1. an airport target detection method based on geographical information space database, it is characterised in that:
The method is divided into two, online and offline processing procedure；Under Xian, according to satellite remote-sensing image and
Geography information, builds the geographical information space of airport target；On line, combining geographic information is empty
The characteristic information of airport target between, it is achieved in unmanned aerial vehicle remote sensing images the detection of airport target with
Location, the method to realize step as follows:
Step S1: combine satellite remote-sensing image and the geographical position of airport target, sets up geography
Information space storehouse；Wherein, the method setting up geographical information space database is, first with man-machine friendship
Mutual mode selects characteristic curve direction, then utilizes the method for line integral to extract airport target and comprises
Characteristic curve, finally utilize characteristic curve attribute information and geography information to build geographical information space
Storehouse；
Step S2: according to positional information and the correction accuracy of unmanned aerial vehicle remote sensing images of unmanned plane
The airport attribute information corresponding with unmanned plane position is extracted, profit from geographical information space database
By the directional information on airport doubtful in geographical information space database, extracted by the method for line integral
Characteristic curve information on unmanned aerial vehicle remote sensing images；
Step S3: by the characteristic curve information on unmanned aerial vehicle remote sensing images and geographical information space database
In the attribute information of doubtful airport feature line carry out Comprehensive Evaluation, determine on unmanned aerial vehicle remote sensing images
The position of characteristic curve whether be the airport position corresponding with unmanned plane position；
Step S4: according to the attribute information of airport target in geographical information space database and characteristic curve
Positional information generate initial part airport profile, thus realize machine in unmanned aerial vehicle remote sensing images
The real-time detection of field target.
A kind of airport target based on geographical information space database the most according to claim 1
Detection method, it is characterised in that: step S2 extracts unmanned plane by the method for line integral
Comprising the concrete steps that of characteristic curve on remote sensing image:
Step 21: according to the geographical coordinate of unmanned aerial vehicle remote sensing images in geographical information space database
Search doubtful airport target；
Step 22: utilize first order differential operator to calculate unmanned aerial vehicle remote sensing images vertical gradient and water
Flat direction gradient；
Step 23: whether be perpendicular to airport side by the gradient judging unmanned aerial vehicle remote sensing images
Reach up to maximum to the doubtful point of characteristic curve judging on unmanned aerial vehicle remote sensing images；
Step 24: by the method for line integral, the characteristic curve on statistical machine field direction doubtful
Point number extracts the characteristic curve on unmanned aerial vehicle remote sensing images.
A kind of airport target based on geographical information space database the most according to claim 1
Detection method, it is characterised in that: step S3 determines the characteristic curve on unmanned aerial vehicle remote sensing images
Position be whether that the method for the airport position corresponding with unmanned plane position is: by combining
Composition and division in a proportion is to characteristic curve in unmanned aerial vehicle remote sensing images and the feature of airport target in geographical information space
Line attribute information judges.
A kind of airport target based on geographical information space database the most according to claim 1
Detection method, it is characterised in that generate initial part airport profile described in step S4
Method is: carry in the attribute of airport target and remote sensing image in comprehensive utilization geographical information space
The characteristic curve taken is to sketch out the accurate location on airport.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310485693.3A CN103488801B (en) | 2013-10-17 | 2013-10-17 | A kind of airport target detection method based on geographical information space database |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310485693.3A CN103488801B (en) | 2013-10-17 | 2013-10-17 | A kind of airport target detection method based on geographical information space database |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103488801A CN103488801A (en) | 2014-01-01 |
| CN103488801B true CN103488801B (en) | 2016-09-28 |
Family
ID=49829027
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310485693.3A Active CN103488801B (en) | 2013-10-17 | 2013-10-17 | A kind of airport target detection method based on geographical information space database |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103488801B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104392234B (en) * | 2014-11-27 | 2017-11-07 | 中国人民解放军国防科学技术大学 | A kind of unmanned plane independent landing object detection method based on image FFT symbolic information |
| CN104978580B (en) * | 2015-06-15 | 2018-05-04 | 国网山东省电力公司电力科学研究院 | A kind of insulator recognition methods for unmanned plane inspection transmission line of electricity |
| CN106951567B (en) * | 2017-04-07 | 2020-07-17 | 安徽建筑大学 | Water system space analysis method based on remote sensing technology for Huizhou traditional settlement |
| CN113589328B (en) * | 2021-08-09 | 2024-01-12 | 深圳市电咖测控科技有限公司 | High-precision GNSS positioning device based on multiple GNSS receiving systems |
| CN114998740B (en) * | 2022-06-13 | 2023-07-21 | 中国电子科技集团公司第五十四研究所 | Airport linear feature extraction method based on line segment distribution |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101153908A (en) * | 2006-09-26 | 2008-04-02 | 中国疾病预防控制中心寄生虫病预防控制所 | System and method for acquiring space and geographical environment information |
| CN102034103A (en) * | 2010-12-03 | 2011-04-27 | 中国科学院软件研究所 | Lineament extraction method of remote sensing image |
| CN103152574A (en) * | 2013-03-15 | 2013-06-12 | 清华大学 | Remote sensing image transmitting system with limited velocity based on feature registration |
-
2013
- 2013-10-17 CN CN201310485693.3A patent/CN103488801B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101153908A (en) * | 2006-09-26 | 2008-04-02 | 中国疾病预防控制中心寄生虫病预防控制所 | System and method for acquiring space and geographical environment information |
| CN102034103A (en) * | 2010-12-03 | 2011-04-27 | 中国科学院软件研究所 | Lineament extraction method of remote sensing image |
| CN103152574A (en) * | 2013-03-15 | 2013-06-12 | 清华大学 | Remote sensing image transmitting system with limited velocity based on feature registration |
Non-Patent Citations (3)
| Title |
|---|
| 一种直线特征遥感影像与GIS矢量数据匹配方法;左经纬等;《测绘科学技术学报》;20101031;第27卷(第5期);352-356 * |
| 利用多光谱遥感图像融合的机场识别方法;隋东坡等;《计算机测量与控制》;20050131(第1期);72-75 * |
| 地理信息库支持下的SAR影像信息提取;王超等;《遥感技术与应用》;20050228;第20卷(第1期);111-115 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103488801A (en) | 2014-01-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104197928B (en) | Multi-camera collaboration-based method for detecting, positioning and tracking unmanned aerial vehicle | |
| CN104536009B (en) | Above ground structure identification that a kind of laser infrared is compound and air navigation aid | |
| CN110595476B (en) | Unmanned aerial vehicle landing navigation method and device based on GPS and image visual fusion | |
| CN106291278B (en) | A kind of partial discharge of switchgear automatic testing method based on more vision systems | |
| CN103488801B (en) | A kind of airport target detection method based on geographical information space database | |
| CN106295512B (en) | Vision data base construction method and indoor orientation method in more correction lines room based on mark | |
| CN102073846B (en) | Method for acquiring traffic information based on aerial images | |
| WO2017076928A1 (en) | Method, device and assembly for map generation | |
| KR101261409B1 (en) | System for recognizing road markings of image | |
| US20190197908A1 (en) | Methods and systems for improving the precision of autonomous landings by drone aircraft on landing targets | |
| CN103605978A (en) | Urban illegal building identification system and method based on three-dimensional live-action data | |
| CN104504675B (en) | A kind of active vision localization method | |
| CN105307115A (en) | Distributed vision positioning system and method based on action robot | |
| CN111079589B (en) | Automatic height detection method based on depth camera shooting and height threshold value pixel calibration | |
| CN107886541B (en) | Real-time monocular moving target pose measuring method based on back projection method | |
| CN111932565A (en) | Multi-target identification tracking resolving method | |
| CN103455815A (en) | Self-adaptive license plate character segmentation method in complex scene | |
| CN105307116A (en) | Distributed vision positioning system and method based on mobile robot | |
| CN108416798A (en) | A kind of vehicle distances method of estimation based on light stream | |
| CN105302132A (en) | Distributed-type vision localization system and method based on travel robot | |
| CN108225273A (en) | A kind of real-time runway detection method based on sensor priori | |
| CN103456027B (en) | Time sensitivity target detection positioning method under airport space relation constraint | |
| US20130093880A1 (en) | Height Measurement Apparatus And Method | |
| CN106709432B (en) | Human head detection counting method based on binocular stereo vision | |
| CN104567879A (en) | Method for extracting geocentric direction of combined view field navigation sensor |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |