CN105678230A - Airplane region of interest spectrum measuring method under the guidance of infrared target projection model - Google Patents
Airplane region of interest spectrum measuring method under the guidance of infrared target projection model Download PDFInfo
- Publication number
- CN105678230A CN105678230A CN201511020660.7A CN201511020660A CN105678230A CN 105678230 A CN105678230 A CN 105678230A CN 201511020660 A CN201511020660 A CN 201511020660A CN 105678230 A CN105678230 A CN 105678230A
- Authority
- CN
- China
- Prior art keywords
- interest
- target
- aircraft
- area
- region
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/64—Three-dimensional objects
- G06V20/653—Three-dimensional objects by matching three-dimensional models, e.g. conformal mapping of Riemann surfaces
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/752—Contour matching
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Medical Informatics (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Databases & Information Systems (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- Image Processing (AREA)
Abstract
The invention discloses an airplane region of interest spectrum measuring method under the guidance of an infrared target projection model. The method includes the steps: 1) establishing a projection posture database; and 2) performing spectrum measuring on a target infrared image of an airplane: including (2.1) acquiring the target infrared image of the airplane through practical measurement; (2.2) acquiring the contour of each region of interest of the target; (2.3) comparing the projection posture database with the practically measured contour of each region of interest of the target, and determining the posture of the airplane and the corresponding geometrical relationship of each region of interest in the posture; (2.4) according to the mutual relation between the overall contour and each region of interest, determining the specific coordinates of each region of interest; (2.5) moving the spectrum measuring central point to the center of each region of interest, and measuring the spectrum of each region of interest one by one; and (2.6) recording the target spectrum features. As the detection model of the airplane region of interest spectrum measuring method under the guidance of an infrared target projection model can conveniently use part of coupling technology, the feature blocks which are not shielded can still be found even if part of the target is shielded and the robustness for detection can be improved.
Description
Technical field
The invention belongs to image detection recognition technology field, more specifically, it relates to the aircraft that a kind of infrared target projection model instructs district interested surveys spectral method, can be used for the locating and tracking in aircraft target district interested and spectral measurement thereof.
Background technology
The radiation at each position of aircraft target is different with reflection spectrum, in order to survey radiation and the reflection spectrum at each position in outfield, it is necessary to detection identifies each area-of-interest of location automatically, and lasting tracking surveys its spectrum. And existing testing method all needs artificial congnition, detection and location area-of-interest, take time and effort very much. It is therefore necessary to development identifies the survey spectral method of location automatically fast.
General aircraft object detection method is all that the global characteristics according to aircraft detects. Given image detects interesting target can be summed up as in yardstick and image space to calculate the probability that interesting target occurs in detection window mouth. Difference according to aircraft target method for expressing, existing interesting target detection method roughly can be divided into two classes: (1) does not use model, only represents interesting target by the feature of some bottoms, is called model-free methods; (2) by the model representation aircraft target designed in advance, it is called there is model method.
Summary of the invention
The general algorithm of target detection based on the overall situation according to target general profile feature, can not can only learn to more meticulous interesting target provincial characteristics. Tradition model is single target model simultaneously, can not represent temperature and the position feature of target region of interest well, the target detect that attitudes vibration is very big is existed certain difficulty. Tradition model is expanded by present method for this reason, and the projection overall situation model of study interesting target and more meticulous projection area-of-interest model improve accuracy of detection.
For achieving the above object, the present invention provides aircraft district's survey interested spectral method that a kind of infrared target projection model instructs, and comprises the steps:
(1) projection attitude data storehouse is set up;
(2) carry out surveying spectrum to actual measurement aircraft target infrared image, comprise following sub-step:
(2.1) actual measurement obtains aircraft target infrared image;
(2.2) infrared image is carried out Iamge Segmentation and edge extraction, thus obtains each area-of-interest profile of target;
(2.3) each area-of-interest profile of target surveyed in the projection attitude data storehouse set up in step (1) and step (2.2) is compared, it is determined that the corresponding geometric relationship of the attitude of aircraft and each area-of-interest under this kind of attitude;
(2.4) according to the mutual relationship between overall profile and each area-of-interest, it is determined that concrete (x, y) coordinate of each area-of-interest;
(2.5) survey spectrum central point is moved to each region of interest centers, measure the spectrum of each area-of-interest one by one;
(2.6) target optical spectrum feature is recorded.
In one embodiment of the present of invention, described step (1) specifically comprises following sub-step:
(1.1) aircraft and the three-dimensional model of tail flame is set up;
(1.2) determine the position of aircraft target unlike material, and unlike material position in three-dimensional model is divided, be specifically divided into the area-of-interests such as head, fuselage, wing edge, passenger cabin, engine and tail flame;
(1.3) different temperature is given to the position of unlike material in three-dimensional model;
(1.4) selected m of the attitude that aloft may occur according to aircraft observes direction, and each is observed direction and is divided into n viewing angle, vertical aircraft m*n projection model of building together;
(1.5) obtain different observation angle to get off the plane the projection model of target;
(1.6) geometric relationship of each area-of-interest that the different observation angle of foundation is got off the plane in aircraft each typical case attitude, area-of-interest comprises head, fuselage, wing edge, passenger cabin, engine and tail flame etc.;
(1.7) set up different observation angle to get off the plane the database of the geometric relationship of each area-of-interest in aircraft each typical case attitude.
In one embodiment of the present of invention, observe direction and be specially 6 and observe directions for described m, be respectively depending on, under depending on, forward sight, after look depending on, left view, the right side.
In one embodiment of the present of invention, described n viewing angle is 9 viewing angles, be respectively face, 15 degree to the left, 30 degree to the left, partially right 15 degree, partially 30 degree, 15 degree on the upper side, 30 degree on the upper side, 15 degree on the lower side, 30 degree on the lower side, the right side.
Compared with prior art, the inventive method has following useful effect:
Present method proposes by the thinking of target global information with local information fusion is improved target detection performance, it is proposed to then first detection overall situation target detects the detection thinking of local region of interest in overall situation model. In the process of detection local region of interest, adopting the target of feature based block to represent, this kind of model does not regard an entirety as target, but target is represented the set for some characteristic blocks. Meanwhile, there is no fixing relative position relation between the characteristic block of present method detection model, have more handiness. In addition, present method detection model can easily use part coupling technology, even if still can find the characteristic block not being blocked when target is at least partially obscured, it is possible to improves detection robustness.
Accompanying drawing explanation
Fig. 1 is that the aircraft district interested that infrared target projection model of the present invention instructs surveys spectral method overall procedure schematic diagram;
Fig. 2 is that the present invention is by the collection of illustrative plates integration apparatus structure block diagram in infrared target locations of contours aircraft district interested;
Fig. 3 is the three-dimensional model of F22 air fighter fuselage and tail flame;
Fig. 4 is the result that F22 carries out material division;
Fig. 5 looks under being and observes direction gained F22 projection model;
Fig. 6 is that forward sight observes direction gained F22 projection model;
Fig. 7 is that left view observes direction gained F22 projection model;
Fig. 8 is that observation direction gained F22 projection model is looked on the right side;
Fig. 9 looks after being and observes direction gained F22 projection model;
Figure 10 looks observation direction gained F22 projection model on being;
Figure 11 for observing under direction upper looking, F22 pitching to the left 15 degree, forward pitch 15 degree, the infrared image of simulation actual measurement;
Figure 12 is the result that the F22 infrared image to simulation actual measurement carries out Iamge Segmentation;
Figure 13 is the F22 infrared image each district interested centre of form mark result to simulation actual measurement;
Figure 14 is the F22 infrared image each district interested positioning result to simulation actual measurement.
Embodiment
In order to make the object of the present invention, technical scheme and advantage clearly understand, below in conjunction with drawings and Examples, the present invention is further elaborated. It is to be understood that specific embodiment described herein is only in order to explain the present invention, it is not intended to limit the present invention. In addition, if below in described each enforcement mode of the present invention involved technology feature do not form conflict each other and just can mutually combine.
As shown in Figure 1, the present invention provides aircraft district's survey interested spectral method that a kind of infrared target projection model instructs, the inventive method is the novel method utilizing collection of illustrative plates integration apparatus quick and precisely to measure, this device structure block diagram as shown in Figure 2, mainly comprises servosystem, camera lens, visual light imaging module, surveys spectrum module.
The inventive method comprises the step setting up projection attitude data storehouse and actual measurement aircraft target infrared image carries out surveying the step of spectrum, wherein:
(1) set up the step in projection attitude data storehouse, specifically comprise following sub-step:
(1.1) setting up the three-dimensional model of F22 air fighter fuselage and tail flame, model is as shown in Figure 3;
(1.2) determine the position of aircraft target unlike material, and unlike material position in three-dimensional model is divided, be specifically divided into: head, fuselage, wing edge, passenger cabin, engine, tail flame. Model partition result is as shown in Figure 4;
(1.3) different temperature is given to the position of unlike material in three-dimensional model;
(1.4) selected 6 of the attitude that may occur according to aerial aircraft observes directions, be respectively depending on, under depending on, forward sight, after look depending on, left view, the right side. Each observe direction be divided into again 9 viewing angles, be respectively face, 15 degree to the left, 30 degree to the left, partially the right side 15 degree, partially the right side 30 degree, 15 degree on the upper side, 30 degree on the upper side, 15 degree on the lower side, 30 degree on the lower side. The overall model of 6*9, vertical aircraft of building together;
(1.5) obtain different observation angle to get off the plane the projection model of target, under look and observe direction gained projection model as shown in Figure 5, forward sight observes direction gained projection model as shown in Figure 6, left view observes direction gained projection model as shown in Figure 7, the right side is looked and is observed direction gained projection model as shown in Figure 8, after look and observe direction gained projection model as shown in Figure 9, above look and observe direction gained projection model as shown in Figure 10;
(1.6) according to above projection model storehouse, the different observation angle of statistics is got off the plane area-of-interest: head, fuselage, wing edge, passenger cabin, engine, the geometric relationship in aircraft each typical case attitude such as tail flame;
(1.7) set up different observation angle to get off the plane area-of-interest: head, fuselage, wing edge, passenger cabin, engine, the database of the geometric relationship in aircraft each typical case attitude such as tail flame;
In above-mentioned steps, concrete Modling model data base method is as follows: the model tormulation of the area-of-interest of feature based block. Present method picture structure framework enriches conventional target detection model, it is possible to obtain the target model being made up of the area-of-interest model of projection overall situation model and projection. First from aircraft infrared picture data storehouse, the learning sample needed for each attitude is chosen, the interesting target of the border rectangle that these samples are marked by band forms, comprise target district interested mark simultaneously, so just can train each projection overall situation model and corresponding projection area-of-interest model. The same with Traditional calculating methods, also it is all the multi-stage characteristics of the projection overall situation model carrying out calculating aircraft under the multiple filter of image. Projection area-of-interest model is obtained by target district interested sample training, it is possible to capture more meticulous target region of interest profile. Therefore, this algorithm is more conducive to accurately detecting target. Projection area-of-interest model departure degree tolerance be by a series of submodule plate and between them geometric relationship form, the optimization aim of model had both comprised the matching degree of each projection area-of-interest model, also comprised the geometry deviation between projection area-of-interest model (in the projection area-of-interest model deviation projection overall situation model distance of correct position).
(2) actual measurement aircraft target infrared image carries out surveying the step of spectrum, specifically comprises following sub-step:
(2.1) simulation actual measurement F22 air fighter target infrared image, as shown in figure 11;
(2.2) infrared image carrying out Iamge Segmentation and edge extraction, thus obtains each area-of-interest profile of target, segmentation result is as shown in figure 12;
(2.3) according to the projection attitude data storehouse set up in step one, it is determined that the observation direction of target image on look, the corresponding geometric relationship of each area-of-interest under this kind of attitude in reading database;
(2.4) according to the mutual relationship between overall profile and each area-of-interest, determine the concrete coordinate of each area-of-interest, head regional centroid coordinate is (291, 70), engine region centre of form coordinate is (230, 229), fuselage regions centre of form coordinate is (247, 185), two tail flame regional centroid coordinates are (197, 293) and (216, 298), two wing edge centre of form coordinates are (219, 156) and (287, 178), the each area-of-interest centre of form mark result of target is as shown in figure 13, the each area-of-interest identification positioning result of target is as shown in figure 14,
(2.5) survey spectrum central point is moved to each region of interest centers, measure the spectrum of each area-of-interest one by one;
(2.6) target optical spectrum feature is recorded.
Those skilled in the art will readily understand; the foregoing is only the better embodiment of the present invention; not in order to limit the present invention, all any amendment, equivalent replacement and improvement etc. done within the spirit and principles in the present invention, all should be included within protection scope of the present invention.
Claims (4)
1. aircraft district's survey interested spectral method that an infrared target projection model instructs, it is characterised in that, described method comprises the steps:
(1) projection attitude data storehouse is set up;
(2) carry out surveying spectrum to actual measurement aircraft target infrared image, comprise following sub-step:
(2.1) actual measurement obtains aircraft target infrared image;
(2.2) infrared image is carried out Iamge Segmentation and edge extraction, thus obtains each area-of-interest profile of target;
(2.3) each area-of-interest profile of target surveyed in the projection attitude data storehouse set up in step (1) and step (2.2) is compared, it is determined that the corresponding geometric relationship of the attitude of aircraft and each area-of-interest under this kind of attitude;
(2.4) according to the mutual relationship between overall profile and each area-of-interest, it is determined that the concrete coordinate of each area-of-interest;
(2.5) survey spectrum central point is moved to each region of interest centers, measure the spectrum of each area-of-interest one by one;
(2.6) target optical spectrum feature is recorded.
2. the method for claim 1, it is characterised in that, described step (1) specifically comprises following sub-step:
(1.1) aircraft and the three-dimensional model of tail flame is set up;
(1.2) determine the position of aircraft target unlike material, and unlike material position in three-dimensional model is divided, be specifically divided into the area-of-interests such as head, fuselage, wing edge, passenger cabin, engine and tail flame;
(1.3) different temperature is given to the position of unlike material in three-dimensional model;
(1.4) selected m of the attitude that aloft may occur according to aircraft observes direction, and each is observed direction and is divided into n viewing angle, vertical aircraft m*n projection model of building together;
(1.5) obtain different observation angle to get off the plane the projection model of target;
(1.6) geometric relationship of each area-of-interest that the different observation angle of foundation is got off the plane in aircraft each typical case attitude, area-of-interest comprises head, fuselage, wing edge, passenger cabin, engine and tail flame etc.;
(1.7) set up different observation angle to get off the plane the database of the geometric relationship of each area-of-interest in aircraft each typical case attitude.
3. method as claimed in claim 1 or 2, it is characterised in that, observe direction and be specially 6 observation directions for described m, be respectively depending on, under depending on, forward sight, after look depending on, left view, the right side.
4. method as claimed in claim 1 or 2, it is characterised in that, described n viewing angle is 9 viewing angles, be respectively face, 15 degree to the left, 30 degree to the left, partially 15 degree, the right side, partially 30 degree, 15 degree on the upper side, 30 degree on the upper side, 15 degree on the lower side, 30 degree on the lower side, the right side.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201511020660.7A CN105678230B (en) | 2015-12-30 | 2015-12-30 | The aircraft region of interest that a kind of infrared target projection model is instructed surveys spectral method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201511020660.7A CN105678230B (en) | 2015-12-30 | 2015-12-30 | The aircraft region of interest that a kind of infrared target projection model is instructed surveys spectral method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105678230A true CN105678230A (en) | 2016-06-15 |
| CN105678230B CN105678230B (en) | 2017-04-05 |
Family
ID=56189775
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201511020660.7A Active CN105678230B (en) | 2015-12-30 | 2015-12-30 | The aircraft region of interest that a kind of infrared target projection model is instructed surveys spectral method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105678230B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106446799A (en) * | 2016-08-31 | 2017-02-22 | 浙江大华技术股份有限公司 | Thermal imaging target identification method and apparatus |
| CN106706133A (en) * | 2016-12-31 | 2017-05-24 | 华中科技大学 | Spot-like target attitude estimation method and system |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103448913A (en) * | 2013-09-11 | 2013-12-18 | 中国民航大学 | Airplane deicing real-time monitoring device |
| EP2933781A2 (en) * | 2014-03-25 | 2015-10-21 | metaio GmbH | Method and system for representing a virtual object in a view of a real environment |
-
2015
- 2015-12-30 CN CN201511020660.7A patent/CN105678230B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103448913A (en) * | 2013-09-11 | 2013-12-18 | 中国民航大学 | Airplane deicing real-time monitoring device |
| EP2933781A2 (en) * | 2014-03-25 | 2015-10-21 | metaio GmbH | Method and system for representing a virtual object in a view of a real environment |
Non-Patent Citations (1)
| Title |
|---|
| 黄士科等: "基于红外图像的被动测距方法", 《红外与激光工程》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106446799A (en) * | 2016-08-31 | 2017-02-22 | 浙江大华技术股份有限公司 | Thermal imaging target identification method and apparatus |
| CN106706133A (en) * | 2016-12-31 | 2017-05-24 | 华中科技大学 | Spot-like target attitude estimation method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105678230B (en) | 2017-04-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104215239B (en) | Guidance method using vision-based autonomous unmanned plane landing guidance device | |
| CN107514993B (en) | The collecting method and system towards single building modeling based on unmanned plane | |
| CN107202982B (en) | A kind of beacon arrangement and image processing method based on UAV position and orientation calculating | |
| KR101445739B1 (en) | Position Detecting Method of Road Traffic Sign | |
| CN109324337B (en) | Unmanned aerial vehicle route generation and positioning method and device and unmanned aerial vehicle | |
| CN104197928B (en) | Multi-camera collaboration-based method for detecting, positioning and tracking unmanned aerial vehicle | |
| CN102436738B (en) | Traffic monitoring device based on unmanned aerial vehicle (UAV) | |
| US20060239537A1 (en) | Automatic processing of aerial images | |
| KR101771492B1 (en) | Method and system for mapping using UAV and multi-sensor | |
| CN113012292A (en) | AR remote construction monitoring method and system based on unmanned aerial vehicle aerial photography | |
| CN111189433A (en) | Karst peak forest landform parameter measuring method based on unmanned aerial vehicle aerial photography | |
| Crispel et al. | All-sky photogrammetry techniques to georeference a cloud field | |
| Schleiss et al. | VPAIR--Aerial Visual Place Recognition and Localization in Large-scale Outdoor Environments | |
| US20130093880A1 (en) | Height Measurement Apparatus And Method | |
| CN105678230B (en) | The aircraft region of interest that a kind of infrared target projection model is instructed surveys spectral method | |
| CN109345590A (en) | A kind of unmanned plane during flying program ver-ify system and method based on binocular vision | |
| CN105631431B (en) | The aircraft region of interest that a kind of visible ray objective contour model is instructed surveys spectral method | |
| CN108225735A (en) | A kind of precision approach indicator flight check method of view-based access control model | |
| CN116957360A (en) | Space observation and reconstruction method and system based on unmanned aerial vehicle | |
| Douret et al. | A multi-cameras 3d volumetric method for outdoor scenes: a road traffic monitoring application | |
| CN113436276B (en) | Visual relative positioning-based multi-unmanned aerial vehicle formation method | |
| Drouin et al. | Modeling and simulation framework for airborne camera systems | |
| CN115601517A (en) | Rock mass structural plane information acquisition method and device, electronic equipment and storage medium | |
| CN111412898B (en) | Large-area deformation photogrammetry method based on ground-air coupling | |
| CN113358325A (en) | Wind tunnel object throwing position and posture measuring method, device, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |