CN110363128A - Airport runway foreign matter detection method, system and medium based on biological vision - Google Patents
Airport runway foreign matter detection method, system and medium based on biological vision Download PDFInfo
- Publication number
- CN110363128A CN110363128A CN201910600355.7A CN201910600355A CN110363128A CN 110363128 A CN110363128 A CN 110363128A CN 201910600355 A CN201910600355 A CN 201910600355A CN 110363128 A CN110363128 A CN 110363128A
- Authority
- CN
- China
- Prior art keywords
- foreign matter
- notable
- biological vision
- airfield runway
- detecting method
- 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.)
- Pending
Links
Classifications
-
- 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
- G06V10/28—Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a method, a system and a medium for detecting foreign matters on an airport runway based on biological vision. The invention uses the biological vision principle and combines the significance detection model to carry out the preliminary detection of the foreign body, thereby achieving the ideal effect and effectively detecting small targets (such as screws, screw caps, gaskets and the like) with low resolution.
Description
Technical field
The present invention relates to airfield runway foreign material detecting techniques, and in particular to a kind of airfield runway foreign matter based on biological vision
Detection method, system and medium.
Background technique
The landing of taking off of aircraft has extremely harsh requirement to runway, frequently uses in view of aircraft and other support vehicles
Runway leads to damage of the runway constantly by similar hollow, takes off also due to takeoff and landing, wake flow and other vibration causes generate
The foreign matter fallen, damage of the FOD to aircraft always exist, but are not paid attention to similarly by with practical flight accident.Very much
Example shows that the exotic on airport runway surface can be easy to be inhaled into engine, leads to power failure, fragment
It can be deposited in mechanical device, influence the normal operation of the equipment such as undercarriage, wing, not only wrap up and preciousness can be seized
Life, but also along with huge economic loss.
Airfield runway foreign matter FOD (Foreign Object Debris), be refer to may damage aircraft or system certain
External substance, clast or object.The type of FOD is many, typically has: metal device (nut, screw, washer, nail, guarantor
Dangerous silk etc.), machine tool, flight article (personal effects, pen, pencil, button etc.), rubbery chip, plastic products, concrete
Pitch fragment (stone, sand, ice slag etc.), paper products, animals and plants etc..
Currently, most domestic airfield runway checks that work is accomplished manually mainly by road face patrolman, in road face
Runway will be closed when inspection, this makes the flight traffic capacity not only low efficiency, poor reliability, it is influenced by environment and people's fatigue strength,
Artificial investigation is also not exclusively reliable, and occupies valuable runway and use the time.
Takeoff and landing runway foreign matter detection system is according to airport runway length and width, and selection is specifically crucial by inspection road face
Front end detection system equipment is installed in runway side subregion in position, on the pylon of desired height, by multisensor integration
Front monitoring front-end and corresponding servomechanism installation are installed on jointly on runway side fixed frame (or runway lights), front end multi-sensor detection system
System sends foreign matter location information to monitoring system display end, and monitoring system adjusts holder and angle, focal length and light according to position
Circle etc. carries out tracking and monitoring to foreign matter and takes pictures, and after carrying out intellectual analysis processing, sends relevant information to master control system for master
Network analysis processing is controlled, and the foreign substance information that will test uploads to command centre, is simultaneously emitted by warning note.At present in the world
More typically there are the Tarsier system of Britain's exploitation, the FODetect system of Israel's exploitation, Singapore to develop
The FOD Finder system of iFerret system and U.S.'s exploitation, as shown in the figure.Tarsier system, FODetect system, FOD
Finder system using millimetre-wave radar detect based on, means supplemented by video image identification technology detect FOD;iFerret
System carries out the detection of FOD only with video image identification technology.Color is not reacted based on the system of Radar Technology, and base
In the system of video image identification technology reaction can be generated to color and illumination contrast.
Existing FOD detection system mainly uses the radar exploration technique, infrared thermal imaging technique and video image identification skill
Art.It is detected automatically using the dangerous situation that radar system is likely to occur runway damage, foreign matter and intersection.It is realized
Principle is by emitting microwave signal to runway, and the mode for collecting reflection signal is detected and analyzed.Since microwave is believed
It number is pulse structure, therefore the time that receiver can be reached by calculating signal obtain the distance between foreign matter and radar.By
It is shorter in the wavelength of radar sensor, and pulse overlap frequency is high, therefore, can be achieved on using radar sensor high
Distance resolution and reduce ground clutter influence.However, having its limitation: radar pair for runway monitoring using radar system
There is high detection effect in metal object, but insensitive for the non-metallic object of rubber etc;Using radar system
Radar shadown and some other obstacle or fixation means influence caused by testing result can not be avoided completely;Slightly for one
Foreign matter false-alarm caused by the strong signal of small metal object reflection is also difficult to exclude.It is detected using infrared or thermal imaging system different
Object avoids potential conflict, however, infrared or thermal imaging system can only realize detection by the infrared ray that sense object emits,
It is then helpless in the case of the temperature and close environment temperature of object.Relative to environmental background temperature, the heat of wisp
Less, infrared or thermal imaging system is difficult to detect.Especially under the conditions of some extreme abnormal weathers (such as: cold day or
Hot day), infrared or thermal imaging system is by being influenced maximum.In addition, infrared or thermal imaging system institute has lost object at picture
The information such as edge, color, be unfavorable for the confirmation and resolution of foreign matter.Using video camera, is returned and schemed by video camera
Foreign matter is found and identified as signal is handled, however, there is also some problems for this method using video camera: right
Airfield runway environment complicated and changeable does not have extensive adaptability, and the Small object low for resolution ratio is easy missing inspection, can generate
A large amount of false-alarm.The technical bottleneck that problem above detects automatically at takeoff and landing runway foreign matter.
The mankind, can be by eye movement pan image scene, so that interested in scene during identifying piece image
Each point falls on the central fovea of retina, and the notable feature of target is preferably recognized using the high-resolution of central fovea.The mankind this
Kind image recognition processes have distinct initiative and selectivity, it emphasizes attention mechanism, and effect is only to allow those senses emerging
Interesting or significant target enters high-rise visual processes, and the content that scene is included, vision are more efficiently understood convenient for us
Attention mechanism plays indispensable role during image understanding and cognition.
Vision attention generally comprises two kinds of mechanism: one is primary vision is based on, by the bottom-up note of data-driven
Meaning;Another kind is based on high-rise vision, top-down attention relevant to task and knowledge.Currently, about bottom-up
Vision attention research is more, has a large amount of computation model and proposes, but estimated performance still has biggish improvement empty compared with the mankind
Between, and for top-down attention, because it is related to high-rise knowledge, it is not easy to form expression, is studied relatively fewer.It is infused with vision
Based on the cognition physiology and psychophysics experiments conclusion of meaning mechanism, vision similar with human visual perception process is established
Attention computation model predicts people's eyes fixation positions, to effectively extract interesting part, solves the problems, such as " target is where ",
Important relevant information is provided for the advanced cognitive process such as target identification.In addition, active vision relevant for task was searched for
Journey, scene information have directive function for the Selective attention of target.
Subgraph (in the reason of having the attention that can attract us at one rapidly be due to color, orientation, shape at this
Or other pattern features have apparent difference compared with its periphery, to highlight.This attention behavior is entirely data
Driving, the participation without high-rise knowledge.Usually it is this it is bottom-up pay attention to be attributed to vision significance, those have compared with
The image-region of strong vision significance tends to the attention for causing observer.Physiologic Studies show corticocerebral some districts
Ditch is pushed up in domain, such as prefrontal lobe Vitrea eye in superior colliculus and side, and there are certain neural response figures, it is significant to one kind of input stimulus
Property expression, the higher place of activity can more arouse attention.Existing airfield runway foreign matter detecting method is with machine vision
Airfield runway foreign bodies detection is solved the problems, such as with image processing techniques, but how to realize the airfield runway foreign matter based on biological vision
Detection, then be still a key technical problem urgently to be resolved.
Summary of the invention
The technical problem to be solved in the present invention: it in view of the above problems in the prior art, provides a kind of based on biological vision
Airfield runway foreign matter detecting method, system and medium, the present invention use biological vision principle, carry out in conjunction with conspicuousness detection model
The Preliminary detection of foreign matter, achieved the effect that it is more satisfactory, can effective detection resolution low Small object (such as screw, spiral shell
Silk cap, washer etc.).
In order to solve the above-mentioned technical problem, the technical solution adopted by the present invention are as follows:
A kind of airfield runway foreign matter detecting method based on biological vision, implementation steps include:
1) video image I is read;
2) biological vision theory is based on to video image I and extracts notable figure S;
3) binary conversion treatment is carried out to notable figure S;
4) each connected region in notable figure S after determining binary conversion treatment is as the foreign matter detected.
Preferably, binary conversion treatment is carried out to notable figure S in step 3) to specifically refer to for each of notable figure S
Pixel sets 255 for the gray value of the pixel if the gray value of the pixel is greater than preset threshold value T.
It preferably, further include being counted to connected region, and examined connected region quantity as foreign matter after step 4)
The step of quantitation exports.
Preferably, further include the steps that carrying out frame to connected region selects after step 4).
Preferably, the detailed step of step 2) includes:
2.1) Fourier transformation is directly carried out to video image I, the real part of Fourier transformation result will be obtained as amplitude
It composes A (f);The imaginary part of Fourier transformation result will be obtained as phase spectrum P (f);
2.2) logarithmic transformation is carried out for amplitude spectrum A (f) obtain log amplitude spectrum L (f);
2.3) calculate log amplitude spectrum L (f), log amplitude spectrum L (f) through the smoothed out result of local average filter convolution it
Between difference, thus obtain spectrum residual error R (f);
2.4) inverse Fourier transform is carried out to spectrum residual error R (f) and phase spectrum P (f) and obtains the notable figure S of spatial domain.
Preferably, the local average filter in step 2.3) is the local average filter h3 (f) of 3*3, and part is flat
The function expression of equal filter h3 (f) are as follows:
It preferably, further include to inverse Fourier transform knot after carrying out inverse Fourier transform to spectrum residual error R (f) in step 2.4)
Fruit is cooked the step of Gaussian Blur filtering that time standard deviation is 8, finally makees the quasi- difference of deutero-albumose for 8 Gaussian Blur filtering output result
For the notable figure S of final spatial domain.
The present invention also provides a kind of foreign body detection system for airfield runway based on biological vision, including computer equipment, should
Computer equipment is programmed or is configured to execute the step of the aforementioned airfield runway foreign matter detecting method based on biological vision of the present invention
Suddenly be stored with or on the storage medium of the computer equipment be programmed or configure it is aforementioned based on biological vision to execute the present invention
The computer program of airfield runway foreign matter detecting method.
The present invention also provides a kind of computer readable storage medium, it is stored with and is programmed on the computer readable storage medium
Or it configures to execute the computer program of the aforementioned airfield runway foreign matter detecting method based on biological vision of the present invention.
The present invention also provides a kind of foreign body detection system for airfield runway based on biological vision, comprising:
Image reading program unit, for reading video image I;
Notable figure extraction procedure unit extracts notable figure S for being based on biological vision theory to video image I;
Binary conversion treatment program unit, for carrying out binary conversion treatment to notable figure S;
Foreign bodies detection program unit, for determining each connected region in the notable figure S after binary conversion treatment as inspection
The foreign matter measured.
Compared to the prior art, in order to solve the above-mentioned technical problem the present invention has an advantage that, the skill that the present invention uses
Art scheme are as follows: be directed to airfield runway foreign bodies detection, the present invention is based on biological vision by reading video image I, to video image I
Theory extracts notable figure S, carries out binary conversion treatment to notable figure S, each connection in notable figure S after determining binary conversion treatment
Region is as the foreign matter detected, and the present invention is by using biological vision principle, in conjunction with top-down attention and from lower
On vision significance model carry out foreign matter Preliminary detection, achieved the effect that it is more satisfactory, can effectively detection resolution it is low
Small object (such as screw, cap nut, washer etc.).
Detailed description of the invention
Fig. 1 is the basic principle schematic of present invention method.
Fig. 2 is that (there are the interference such as various textures, tag line and greasy dirt in the visible runway face as original image of the embodiment of the present invention
Factor).
Fig. 3 is the notable figure S obtained as the embodiment of the present invention.
Fig. 4 is the detection effect as the embodiment of the present invention.
Fig. 5 is the detection effect of the existing optical flow method compared as the embodiment of the present invention.
Fig. 6 is the detection effect of the existing background subtraction compared as the embodiment of the present invention.
Fig. 7 is the detection effect of the existing frame difference method compared as the embodiment of the present invention.
Fig. 8 is the detection effect of the existing mixed Gauss model method compared as the embodiment of the present invention.
Specific embodiment
As shown in Figure 1, the implementation steps of airfield runway foreign matter detecting method of the present embodiment based on biological vision include:
1) video image I is read;In the present embodiment, detected video image I is as shown in Fig. 2, practical includes 8 airports
Runway foreign matter;
2) biological vision theory is based on to video image I and extracts notable figure S, as shown in Figure 3;
3) binary conversion treatment is carried out to notable figure S;
4) each connected region in notable figure S after determining binary conversion treatment is as the foreign matter detected.
In the present embodiment, binary conversion treatment is carried out to notable figure S in step 3) and is specifically referred to for every in notable figure S
One pixel sets 255 for the gray value of the pixel if the gray value of the pixel is greater than preset threshold value T.
It further include being counted to connected region in the present embodiment, after step 4), and using connected region quantity as different
The step of quality testing quantitation exports.
In the present embodiment, further include the steps that carrying out frame to connected region selects after step 4).
In the present embodiment, the detailed step of step 2) includes:
2.1) Fourier transformation is directly carried out to video image I, the real part of Fourier transformation result will be obtained as amplitude
It composes A (f);The imaginary part of Fourier transformation result will be obtained as phase spectrum P (f);It may be expressed as:
A (f)=R (F [I (x)]), P (f)=H (F [I (x)])
Wherein, F [*] indicates Fourier transformation, and R (*) expression takes real part, and H (*) expression takes imaginary part.
2.2) logarithmic transformation is carried out for amplitude spectrum A (f) obtain log amplitude spectrum L (f), it may be assumed that
L (f)=log (A (f))
2.3) calculate log amplitude spectrum L (f), log amplitude spectrum L (f) through the smoothed out result of local average filter convolution it
Between difference, thus obtain spectrum residual error R (f), it may be assumed that
R (f)=L (f)-h3 (f) * L (f)
Local average filter is used to the amplitude spectrum after approximate mass data is averaged, it can be achieved that log amplitude spectrum L's (f)
The content of smoothing processing, residual error spectrum can also be construed to the unexpected part of image.
2.4) inverse Fourier transform is carried out to spectrum residual error R (f) and phase spectrum P (f) and obtains the notable figure S of spatial domain.
In the present embodiment, the local average filter h3 (f) that the local average filter in step 2.3) is 3*3, and office
The function expression of portion average filter h3 (f) are as follows:
It further include to Fourier's inversion after carrying out inverse Fourier transform to spectrum residual error R (f) in step 2.4) in the present embodiment
The step of result does the Gaussian Blur filtering that time standard deviation is 8 is changed, finally ties deutero-albumose quasi- difference for 8 Gaussian Blur filtering output
Notable figure S of the fruit as final spatial domain.By Fourier inversion, output image, as Saliency maps are constructed in airspace,
Conspicuousness mapping is mainly comprising part and parcel in scene.Inverse Fourier transform is carried out to R (f), obtains the conspicuousness of spatial domain
Figure does the Gaussian Blur that time standard deviation is 8 to Saliency maps and filters, obtains the notable figure S (x) of more preferable visual effect, have:
S (x)=g (x) * F-1[exp(R(f)+P(f))]
In above formula, g (x) is the Gaussian Blur filtering that time standard deviation is 8, F-1[*] is inverse Fourier transform, and R (f) is that spectrum is residual
Difference, P (f) are phase spectrum.
In addition, the present embodiment also provides a kind of foreign body detection system for airfield runway based on biological vision, including computer
Equipment, the computer equipment are programmed or are configured to execute the aforementioned airfield runway foreign bodies detection based on biological vision of the present embodiment
It is stored on the storage medium of the step of method or the computer equipment and is programmed or configures that the present embodiment is aforementioned to be based on to execute
The computer program of the airfield runway foreign matter detecting method of biological vision.In addition, the present embodiment also provide it is a kind of computer-readable
Storage medium, be stored on the computer readable storage medium be programmed or configure it is aforementioned based on biology view to execute the present embodiment
The computer program of the airfield runway foreign matter detecting method of feel.
In addition, the present embodiment also provides a kind of foreign body detection system for airfield runway based on biological vision, comprising:
Image reading program unit, for reading video image I;
Notable figure extraction procedure unit extracts notable figure S for being based on biological vision theory to video image I;
Binary conversion treatment program unit, for carrying out binary conversion treatment to notable figure S;
Foreign bodies detection program unit, for determining each connected region in the notable figure S after binary conversion treatment as inspection
The foreign matter measured.
As shown in figure 4, the present embodiment detects 8 foreign matters based on the airfield runway foreign matter detecting method of biological vision altogether, it is real
Border is 8 foreign matters, accuracy in detection 100%.In order to verify airfield runway foreign bodies detection of the present embodiment based on biological vision
The accuracy of method, the same image of the airfield runway foreign matter detecting method for the present embodiment based on biological vision, is adopted respectively
Detection comparison is carried out with existing optical flow method, background subtraction, frame difference method, mixed Gauss model method.
The foreign bodies detection principle of optical flow method is according to the speed of foreign matter moving target pixel transient motion, when real-time detection just
As the light stream of variation, it includes the two-dimensional vector field of each pixel transient motion velocity vector information, and the grade for reacting gray scale becomes
Change, research is structure and its relationship of movement in the variation and scene of the gray scale of image in time.Fig. 5 is for same field
Scape uses the result of the foreign bodies detection of existing optical flow method.Although optical flow method is capable of detecting when foreground moving object, still, light stream
Method it is computationally intensive, operation time is long, if do not improve be not able to satisfy substantially detection real-time requirement, and optical flow method examine
The parameter that moving target lacks target sizes and brightness control is surveyed, noise is obvious.
Background subtraction is also referred to as background subtracting method, and principle is fairly simple, will test target as prospect, and other are unrelated
Target first passes through some algorithms to background modeling as background before detection, then will be after the completion of image to be detected and modeling
Background image is checked the mark, and a threshold value T is set, and is background by the pixel region that difference result is lower than this threshold value, higher than the pixel of threshold value
Region is moving target, is common classic algorithm.Fig. 6 is to be examined for Same Scene using the foreign matter of existing background subtraction
The result of survey.It is one of detection algorithm more commonly used at present to airport foreign bodies detection with background subtraction, its advantage is that calculating
Method realizes that simply complexity is relatively low, therefore can substantially meet the needs of airport foreign bodies detection real-time, and be generally possible to
It obtains than more complete target signature data, the fixed scene of camera is more applicable in.But it is single with background subtraction
It is actual with during will appear many problems, if inconsiderate to these problems complete, the knot of detection will be directly affected
Fruit.
Frame difference method is one of most common method of moving object detection algorithm, algorithm based on time-series image it is adjacent two
Motion target area and background area in frame or multiple image are separated by subtracting operation, and principle is by the corresponding picture of consecutive frame
Element value carries out subtraction, to obtain inter-frame difference image, then pixel is become difference image binaryzation by given threshold
Change value and threshold value comparison, if when bigger than the threshold value set, so that it may which corresponding pixel is determined as foreground pixel, i.e. movement mesh
Mark.If difference image respective pixel is less than the threshold value of setting, which is determined as background pixel point.Adjacent two frame is due to phase
The time of difference is very short, approximate previous frame image can be regarded background image, a later frame image is as present image, therefore background is not
It can accumulate, the speed of update is fast, and algorithm is simple, and calculation amount is small, substantially meets the needs of detection real-time.Fig. 7 is for same
Scene uses the result of the foreign bodies detection of existing frame difference method.There is also shortcomings for adjacent frame differential method, such as in environment
Noise it is very sensitive, be easy by noise erroneous detection be target to be detected, and threshold value selection to determine detection object effects it is larger.
When and target gray larger for target is approximate with background gray scale, it is easy to generate cavity inside moving target, this is to subsequent
The calculating of connected component labeling, connected domain area has an impact.Adjacent frame differential method also makes objective contour imperfect, to movement mesh
Subsequent calibration is marked to be affected.Therefore, although simple using adjacent frame differential method algorithm merely, calculation amount is small, not
Testing requirements can be fully met, it is necessary to improve it and other algorithms are used in combination to enhance testing result.
Mixed Gauss model method analyzes changeable movement by establishing the mixed Gauss model being made of multiple single Gausses
Object, the background modeling being relatively suitble in complex scene.Each pixel brightness value in continuous N frame image is fitted
Normal distribution situation out recycles mean value and variance the two parameters to find out confidence interval, the pixel brightness value conduct in this section
Initial background image.Fig. 8 is the result for Same Scene using the foreign bodies detection of existing mixed Gauss model method.Mixing is high
The advantages of this model algorithm is the moving object detection that can adapt in complex background, the disadvantage is that Gauss model is computationally intensive,
And context update is not prompt enough, leads in scene that there is a situation where when larger illumination variation, testing result is not ideal enough.
5~Fig. 8 of comparison diagram it is found that relatively existing optical flow method, background subtraction, frame difference method, mixed Gauss model method and
Speech, the present embodiment use biological vision principle based on the airfield runway foreign matter detecting method of biological vision, detect in conjunction with conspicuousness
Model carries out the Preliminary detection of foreign matter, achieved the effect that it is more satisfactory, can effective detection resolution low Small object (such as spiral shell
Adjusting screw, cap nut, washer etc.).
It should be understood by those skilled in the art that, embodiments herein can provide as method, system or computer program
Product.Therefore, complete hardware embodiment, complete software embodiment or reality combining software and hardware aspects can be used in the application
Apply the form of example.Moreover, it wherein includes the computer of computer usable program code that the application, which can be used in one or more,
The computer program implemented in usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) produces
The form of product.
The application is referring to method, the process of equipment (system) and computer program product according to the embodiment of the present application
Figure and/or block diagram describe.It should be understood that every one stream in flowchart and/or the block diagram can be realized by computer program instructions
The combination of process and/or box in journey and/or box and flowchart and/or the block diagram.It can provide these computer programs
Instruct the processor of general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produce
A raw machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute for real
The device for the function of being specified in present one or more flows of the flowchart and/or one or more blocks of the block diagram.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates,
Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or
The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one
The step of function of being specified in a box or multiple boxes.
The above is only a preferred embodiment of the present invention, protection scope of the present invention is not limited merely to above-mentioned implementation
Example, all technical solutions belonged under thinking of the present invention all belong to the scope of protection of the present invention.It should be pointed out that for the art
Those of ordinary skill for, several improvements and modifications without departing from the principles of the present invention, these improvements and modifications
It should be regarded as protection scope of the present invention.
Claims (10)
1. a kind of airfield runway foreign matter detecting method based on biological vision, it is characterised in that implementation steps include:
1) video image I is read;
2) biological vision theory is based on to video image I and extracts notable figure S;
3) binary conversion treatment is carried out to notable figure S;
4) each connected region in notable figure S after determining binary conversion treatment is as the foreign matter detected.
2. the airfield runway foreign matter detecting method according to claim 1 based on biological vision, which is characterized in that step 3)
In binary conversion treatment carried out to notable figure S specifically refer to for each of notable figure S pixel, if the gray scale of the pixel
Value is greater than preset threshold value T, then sets 255 for the gray value of the pixel.
3. the airfield runway foreign matter detecting method according to claim 1 based on biological vision, which is characterized in that step 4)
Further include later connected region is counted, and using connected region quantity as foreign bodies detection quantity export the step of.
4. the airfield runway foreign matter detecting method according to claim 1 based on biological vision, which is characterized in that step 4)
Further include the steps that carrying out frame to connected region selects later.
5. the airfield runway foreign matter detecting method described according to claim 1~any one of 4 based on biological vision, special
Sign is that the detailed step of step 2) includes:
2.1) Fourier transformation is directly carried out to video image I, the real part of Fourier transformation result will be obtained as amplitude spectrum A
(f);The imaginary part of Fourier transformation result will be obtained as phase spectrum P (f);
2.2) logarithmic transformation is carried out for amplitude spectrum A (f) obtain log amplitude spectrum L (f);
2.3) log amplitude spectrum L (f), log amplitude spectrum L (f) are calculated through between the smoothed out result of local average filter convolution
Difference, to obtain spectrum residual error R (f);
2.4) inverse Fourier transform is carried out to spectrum residual error R (f) and phase spectrum P (f) and obtains the notable figure S of spatial domain.
6. the airfield runway foreign matter detecting method according to claim 5 based on biological vision, which is characterized in that step
2.3) the local average filter h3 (f) that the local average filter in is 3*3, and the function of local average filter h3 (f)
Expression formula are as follows:
7. the airfield runway foreign matter detecting method according to claim 5 based on biological vision, which is characterized in that step
It 2.4) further include the height for doing time standard deviation to inverse Fourier transform result and being 8 after carrying out inverse Fourier transform to spectrum residual error R (f) in
The quasi- difference of deutero-albumose is finally exported result as the aobvious of final spatial domain for 8 Gaussian Blur filtering by the step of this fuzzy filter
Write figure S.
8. a kind of foreign body detection system for airfield runway based on biological vision, including computer equipment, which is characterized in that the calculating
Machine equipment is programmed or is configured with the airfield runway foreign matter based on biological vision described in any one of perform claim requirement 1~7
It is stored with to be programmed or configure on the storage medium of the step of detection method or the computer equipment and requires 1~7 with perform claim
Any one of described in the airfield runway foreign matter detecting method based on biological vision computer program.
9. a kind of computer readable storage medium, which is characterized in that be stored on the computer readable storage medium be programmed or
Configuration is with the calculating of the airfield runway foreign matter detecting method based on biological vision described in any one of perform claim requirement 1~7
Machine program.
10. a kind of foreign body detection system for airfield runway based on biological vision, characterized by comprising:
Image reading program unit, for reading video image I;
Notable figure extraction procedure unit extracts notable figure S for being based on biological vision theory to video image I;
Binary conversion treatment program unit, for carrying out binary conversion treatment to notable figure S;
Foreign bodies detection program unit, for determining each connected region in the notable figure S after binary conversion treatment as detecting
Foreign matter.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910600355.7A CN110363128A (en) | 2019-07-04 | 2019-07-04 | Airport runway foreign matter detection method, system and medium based on biological vision |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910600355.7A CN110363128A (en) | 2019-07-04 | 2019-07-04 | Airport runway foreign matter detection method, system and medium based on biological vision |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110363128A true CN110363128A (en) | 2019-10-22 |
Family
ID=68218066
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910600355.7A Pending CN110363128A (en) | 2019-07-04 | 2019-07-04 | Airport runway foreign matter detection method, system and medium based on biological vision |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110363128A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112836587A (en) * | 2021-01-08 | 2021-05-25 | 中国商用飞机有限责任公司北京民用飞机技术研究中心 | Runway identification method and device, computer equipment and storage medium |
CN113808028A (en) * | 2020-09-14 | 2021-12-17 | 北京航空航天大学 | Attribution algorithm-based confrontation sample detection method and device |
CN112836587B (en) * | 2021-01-08 | 2024-06-04 | 中国商用飞机有限责任公司北京民用飞机技术研究中心 | Runway identification method, runway identification device, computer equipment and storage medium |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103149563A (en) * | 2013-01-28 | 2013-06-12 | 北京航空航天大学 | Two-dimension detection system and method for airport pavement foreign matter |
CN104021557A (en) * | 2014-06-13 | 2014-09-03 | 沈阳飞机工业(集团)有限公司 | Airport near-space complex environment foreign matter monitoring and early warning method |
CN106845346A (en) * | 2016-12-16 | 2017-06-13 | 北京无线电计量测试研究所 | A kind of image detecting method for airfield runway foreign bodies detection |
CN108665446A (en) * | 2018-04-17 | 2018-10-16 | 上海工程技术大学 | A kind of foreign body detection system for airfield runway and method with radar |
CN109635649A (en) * | 2018-11-05 | 2019-04-16 | 航天时代飞鸿技术有限公司 | A kind of high speed detection method and system of unmanned plane spot |
-
2019
- 2019-07-04 CN CN201910600355.7A patent/CN110363128A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103149563A (en) * | 2013-01-28 | 2013-06-12 | 北京航空航天大学 | Two-dimension detection system and method for airport pavement foreign matter |
CN104021557A (en) * | 2014-06-13 | 2014-09-03 | 沈阳飞机工业(集团)有限公司 | Airport near-space complex environment foreign matter monitoring and early warning method |
CN106845346A (en) * | 2016-12-16 | 2017-06-13 | 北京无线电计量测试研究所 | A kind of image detecting method for airfield runway foreign bodies detection |
CN108665446A (en) * | 2018-04-17 | 2018-10-16 | 上海工程技术大学 | A kind of foreign body detection system for airfield runway and method with radar |
CN109635649A (en) * | 2018-11-05 | 2019-04-16 | 航天时代飞鸿技术有限公司 | A kind of high speed detection method and system of unmanned plane spot |
Non-Patent Citations (2)
Title |
---|
党国龙: "《机场跑道异物检测与识别的相关算法研究》", 《中国优秀硕士学位论文全文数据库工程科技Ⅰ辑》 * |
田赛: "《视频显著区域检测研究》", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113808028A (en) * | 2020-09-14 | 2021-12-17 | 北京航空航天大学 | Attribution algorithm-based confrontation sample detection method and device |
CN113808028B (en) * | 2020-09-14 | 2023-08-08 | 北京航空航天大学 | Method and device for detecting countermeasure sample based on attribution algorithm |
CN112836587A (en) * | 2021-01-08 | 2021-05-25 | 中国商用飞机有限责任公司北京民用飞机技术研究中心 | Runway identification method and device, computer equipment and storage medium |
CN112836587B (en) * | 2021-01-08 | 2024-06-04 | 中国商用飞机有限责任公司北京民用飞机技术研究中心 | Runway identification method, runway identification device, computer equipment and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Wang et al. | A comparative study of state-of-the-art deep learning algorithms for vehicle detection | |
CN107609522B (en) | Information fusion vehicle detection system based on laser radar and machine vision | |
CN111626290B (en) | Infrared ship target detection and identification method under complex sea surface environment | |
Lai et al. | Characterization of Sky‐region Morphological‐temporal Airborne Collision Detection | |
Dai et al. | Multi-task faster R-CNN for nighttime pedestrian detection and distance estimation | |
WO2010084902A1 (en) | Intrusion alarm video processing device | |
CN111899568B (en) | Bridge anti-collision early warning system, method and device and storage medium | |
CN106372619B (en) | A kind of detection of vehicle robust and divided lane reach summation curve estimation method | |
CN112800860A (en) | Event camera and visual camera cooperative high-speed scattered object detection method and system | |
CN110674674A (en) | Rotary target detection method based on YOLO V3 | |
CN111913177A (en) | Method and device for detecting target object and storage medium | |
CN111323756A (en) | Deep learning-based marine radar target detection method and device | |
CN111323757B (en) | Target detection method and device for marine radar | |
Lyu et al. | Sea-surface object detection based on electro-optical sensors: A review | |
Ihekoronye et al. | Aerial supervision of drones and other flying objects using convolutional neural networks | |
CN114898319A (en) | Vehicle type recognition method and system based on multi-sensor decision-level information fusion | |
CN110363128A (en) | Airport runway foreign matter detection method, system and medium based on biological vision | |
Sezgin et al. | Safe autonomous driving in adverse weather: Sensor evaluation and performance monitoring | |
CN113189590A (en) | FOD-based target identification method and device and storage medium | |
Bi et al. | Machine vision | |
CN115965847A (en) | Three-dimensional target detection method and system based on multi-modal feature fusion under cross view angle | |
CN109493366A (en) | Based on background modeling method, system and the medium for visiting bird radar image | |
Li et al. | Research on detection technology of autonomous landing based on airborne vision | |
Huang et al. | A radar and monocular camera-based fusion approach for pedestrian detection | |
CA3138095A1 (en) | Computer vision aircraft detection |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191022 |
|
RJ01 | Rejection of invention patent application after publication |