CN107481233A - A kind of image-recognizing method being applied in FOD foreign bodies detection radars - Google Patents
A kind of image-recognizing method being applied in FOD foreign bodies detection radars Download PDFInfo
- Publication number
- CN107481233A CN107481233A CN201710724260.7A CN201710724260A CN107481233A CN 107481233 A CN107481233 A CN 107481233A CN 201710724260 A CN201710724260 A CN 201710724260A CN 107481233 A CN107481233 A CN 107481233A
- Authority
- CN
- China
- Prior art keywords
- image
- detection
- fod
- processing
- foreign matter
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10032—Satellite or aerial image; Remote sensing
- G06T2207/10044—Radar image
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Image Processing (AREA)
Abstract
The present invention provides a kind of image-recognizing method being applied in FOD foreign bodies detection radars, including：Step 1：Straight-line detection is carried out to the original image of collection, to remove the image at airfield runway edge；Step 2：Image preprocessing is carried out to the image after step 1 processing, so that feature interested in image selectively protrudes, and the unwanted feature that decays；Step 3：Plaque detection is carried out to the image after step 2 processing, foreign matter is confirmed whether it is by patch, and testing result is fed back.The present invention uses image recognition technology, disturbing factor in the image acquired is excluded by means such as straight-line detection, image preprocessings, then the method for plaque detection is passed through, microsize (2cm) FOD in image is positioned, and judge the affiliated species of FOD, most processing result image reports airfield runway maintenance centre at last, so as to improve the security of Airport Operation.
Description
Technical field
The present invention relates to technical field of aerospace, more particularly to a kind of image recognition being applied in FOD foreign bodies detection radars
Method.
Background technology
In aerospace field, foreign matter (Foreign Object Debris, the abbreviation to aircraft runway are often needed
FOD) detected, be existing frequently-used and effective technology, but FOD radar systems by FOD foreign bodies detection detections of radar FOD
The image got is relatively fuzzyyer, and it, which sends to contain in the real image in airfield runway maintenance centre, is not detected by volume
Smaller, imaging Relative Fuzzy FOD.
The content of the invention
It is an object of the invention to solve the defects of above-mentioned prior art is present, there is provided one kind is able to detect that aircraft is run
The smaller FOD of road upper volume image processing method.
A kind of image-recognizing method being applied in FOD foreign bodies detection radars, comprises the following steps：
Step 1：Straight-line detection is carried out to the original image of collection, to remove the image at airfield runway edge；
Step 2：Image preprocessing is carried out to the image after step 1 processing, so that feature interested in image has selection
Protrusion, and the unwanted feature that decays；
Step 3：Plaque detection is carried out to the image after step 2 processing, foreign matter is confirmed whether it is by patch, and will detection
As a result fed back.
Further, method as described above, before step 3 testing result is fed back, in addition to detecting
Foreign matter is classified according to shape size, and classification and Detection result is fed back.
Further, method as described above, before step 1, in addition to：Pass through camera persistent collection image first
Data, when FOD radars find foreign matter, the infrared foreign matter place orientation that can be transferred to is so that camera shoots the image become apparent from
Original image of the data as step 1.
Further, method as described above, the straight-line detection use sub-pixel feature location algorithm.
Further, method as described above, described image pretreatment are handled image using frequency domain method.
Further, method as described above, the plaque detection include dividing the image after previous step processing
The partitioning algorithm processing of water ridge, the gradient image obtained after fractional spins are handled carries out threshold process, to eliminate ash
Over-segmentation caused by the minor variations of degree.
Beneficial effect：
The present invention uses image recognition technology, and the disturbing factor in the image acquired is passed through into straight-line detection, image
The means such as pretreatment are excluded, and then by the method for plaque detection, microsize (2cm) FOD in image is positioned, and judges
The affiliated species of FOD, most processing result image reports airfield runway maintenance centre at last, so as to improve the safety of Airport Operation
Property.
Brief description of the drawings
Fig. 1 is image-recognizing method flow chart of the present invention.
Embodiment
To make the object, technical solutions and advantages of the present invention clearer, below technical scheme in the present invention carry out it is clear
Chu, it is fully described by, it is clear that described embodiment is part of the embodiment of the present invention, rather than whole embodiments.It is based on
Embodiment in the present invention, those of ordinary skill in the art are obtained every other under the premise of creative work is not made
Embodiment, belong to the scope of protection of the invention.
Optical system airport foreign bodies detection is mainly concerned with straight-line detection, image preprocessing, foreign matter plaque detection etc.；Image
Pretreatment is in order to which the disturbing factors such as weather are rejected from image.The purpose of straight-line detection is to detect the side of airfield runway
Foreign bodies detection and processing are not done in edge, runway outside.Foreign matter shows as a patch on image, so plaque detection algorithm is different
The key of thing extraction.
Fig. 1 is image-recognizing method flow chart of the present invention, as shown in figure 1, method provided by the invention comprises the following steps：
Step 1：Camera persistent collection view data；
Step 2：The data that step 1 is collected are sent to process plate；
Step 3：Whether FOD detections of radar has foreign matter, if without foreign matter, then return to step 2；If detect different
Thing, then it is infrared to be transferred to foreign matter (FOD) place orientation；It is infrared to be transferred to foreign matter (FOD) place orientation, facilitate camera collection more
Add clearly view data, view data can be sent to process plate；
Step 4：Step 3 is gathered into figure and carries out whether straight-line detection has runway exterior domain to calculate, if so, then handling
Plate can calculate runway edge automatically, and the region outside runway is got rid of, subsequently into step once；If there is no runway edge,
Then it is directly entered next step；
Step 5：Image preprocessing is carried out to the figure after progress straight-line detection, so that feature interested in image has choosing
The protrusion selected, and the unwanted feature that decays；
Step 6：Plaque detection is carried out to pretreated image, foreign matter is confirmed whether it is by patch, if not different
Thing, then directly the result detected is fed back, i.e., most at last this clearly target image be transferred to runway maintenance centre clothes
It is engaged on device, or the computer of user in need；If foreign matter, then into step 7；
Step 7：The foreign matter detected is classified according to its shape size, and classification and Detection result is fed back,
I.e. most at last this clearly sorted target image is transferred to runway maintenance centre server, or the electricity of user in need
On brain.
Straight-line detection
The purpose of straight-line detection is to detect the edge of airfield runway, foreign bodies detection and processing are not done outside runway.Directly
Line detection mainly uses the detection method based on edge：
Classical edge detection operator utilizes the gradient extremum characteristic of image border mostly.It is maximum to detect local first derivative
Or second dervative zero crossing is as marginal point.Such as the methods of Roberts, Sobel and Laplacian operator, due to directly entering
Row is differentiated, so their anti-noise jamming ability is than relatively low.Grow up on the basis of the edge detection operator of classics
LOG (Laplacian of gauss) operator, Canny operators, wavelet transform dimension edge detection method is all different degrees of
On original edge detection method is improved, but the positioning precision of its rim detection is typically only capable to reach Pixel-level.For
Further raising positioning precision, using sub-pixel feature location algorithm, the algorithm is earliest due to playing peak in the essence based on image
It is a variety of to have developed into interpolation fitting method, spatial gradation method and digital correlation registration method etc. with being proposed in motion measurement for close measurement
Detection method.
Image preprocessing
The factor for influenceing image definition has a lot, and when particularly night uses infrared illumination lamp, illuminance is not uniform enough
When will result in gradation of image and excessively concentrate, so, it is necessary to carry out strengthening the pre- places such as noise reduction to digital picture before image detection
Reason.By CCD (camera) obtain image by A/D (D/A switch, the function in picture system by Data Acquisition Card Lai
Realize) conversion, circuit transmission can all produce noise pollution etc..Therefore picture quality inevitably reduces, the lighter's performance
It is unclean for image, it is difficult to see details clearly, severe one shows as that image is smudgy, and general picture also be can't see.Therefore, employ
The method of image enhaucament：Image enhaucament does not consider the reason for image quality decrease, and feature interested in image only is had into selection
Protrusion, and the unwanted feature that decays, its purpose is mainly to improve the intelligibility of image.Wherein, image enhaucament employs
Frequency domain method, in some transform domain of image, image is operated, the coefficient after modification conversion, such as Fourier transformation,
The coefficient of dct transform etc., the image after inverse transformation is handled then is carried out again.In use, airport is certain to system
There is different weather conditions, such as rain, mist etc., in addition the illumination variation in daytime and evening, make the clear of image at different moments
Degree changes.
Plaque detection
Foreign matter shows as a patch on image, so plaque detection algorithm is the key of foreign matter extraction.Plaque detection
Mainly use fractional spins：
The calculating process in watershed is an iteration annotation process.Calculate in two steps, one is sequencer procedure, one
It is the process of flooding.The gray level of each pixel is sorted from low to high first, then floods process in realization from low to high
In, the domain of influence of each local minimum in h rank height is judged and marked using first in first out (FIFO) structure.
What watershed transform obtained is the reception basin image of input picture, the boundary point between reception basin, as watershed.
Obviously, what watershed represented is input picture maximum point.Therefore, to obtain the marginal information of image, generally gradient image
As input picture, i.e.,
G (x, y)=grad (f (x, y))={ [f (x, y)-f (x-1, y)] 2 [f (x, y)-f (x, y-1)] 2 } 0.5
In formula, f (x, y) represents original image, and grad { } represents gradient algorithm.
Because gradient function can preferably describe border, the application is because gradient as input picture using gradient image
Image can preferably show boundary point, and gradient image is got by carrying out grad { } gradient algorithm to original image.
Watershed algorithm has good response to faint edge, and the trickle gray scale of the noise, body surface in image becomes
Change, can all produce the phenomenon of over-segmentation.But simultaneously it should be observed that watershed algorithm has good response to faint edge,
It is to obtain the guarantee of closing continuous boundary.In addition, the reception basin of the closing obtained by watershed algorithm, to analyze the area of image
Characteristic of field provides possibility.
To eliminate over-segmentation problem caused by watershed algorithm, the present invention causes reception basin only using modification gradient function
The desired target detected is responded to solve the problem.
For over-segmentation caused by reduction watershed algorithm, generally gradient function is modified, a simple side
Method is to carry out threshold process to gradient image, to eliminate over-segmentation caused by the minor variations of gray scale.I.e.
G (x, y)=max (grad (f (x, y)), g θ)
G θ represent threshold value.
Program can use method：Eliminated with threshold restriction gradient image with reaching caused by the minor variations of gray value excessively
Segmentation, obtains appropriate region, then the gray level of the marginal point in these regions is sorted from low to high, then from it is low to
Height realizes the process flooded, and gradient image is calculated with Sobel operators and obtained.When carrying out threshold process to gradient image, choose and close
Suitable threshold value has a significant impact to the image finally split, therefore the selection of threshold value is a pass of image segmentation quality
Key.
Finally it should be noted that：The above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although
The present invention is described in detail with reference to the foregoing embodiments, it will be understood by those within the art that：It still may be used
To be modified to the technical scheme described in foregoing embodiments, or equivalent substitution is carried out to which part technical characteristic；
And these modification or replace, do not make appropriate technical solution essence depart from various embodiments of the present invention technical scheme spirit and
Scope.
Claims (6)
1. a kind of image-recognizing method being applied in FOD foreign bodies detection radars, it is characterised in that comprise the following steps：
Step 1：Straight-line detection is carried out to the original image of collection, to remove the image at airfield runway edge；
Step 2：Image preprocessing is carried out to the image after step 1 processing, so that feature interested in image is selectively dashed forward
Go out, and the unwanted feature that decays；
Step 3：Plaque detection is carried out to the image after step 2 processing, foreign matter is confirmed whether it is by patch, and by testing result
Fed back.
2. according to the method for claim 1, it is characterised in that before step 3 testing result is fed back, in addition to pair
The foreign matter detected is classified according to shape size, and classification and Detection result is fed back.
3. according to the method for claim 1, it is characterised in that before step 1, in addition to：Held first by camera
Continuous to collect view data, when FOD radars find foreign matter, the infrared foreign matter place orientation that can be transferred to is so that camera is shot more
Clearly original image of the view data as step 1.
4. according to the method for claim 1, it is characterised in that the straight-line detection uses sub-pixel feature location algorithm.
5. according to the method for claim 1, it is characterised in that described image pretreatment using frequency domain method to image at
Reason.
6. according to the method for claim 1, it is characterised in that the plaque detection is included to the figure after previous step processing
As carrying out fractional spins processing, the gradient image obtained after fractional spins are handled carries out threshold process,
To eliminate over-segmentation caused by the minor variations of gray scale.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710724260.7A CN107481233A (en) | 2017-08-22 | 2017-08-22 | A kind of image-recognizing method being applied in FOD foreign bodies detection radars |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710724260.7A CN107481233A (en) | 2017-08-22 | 2017-08-22 | A kind of image-recognizing method being applied in FOD foreign bodies detection radars |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107481233A true CN107481233A (en) | 2017-12-15 |
Family
ID=60602230
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710724260.7A Pending CN107481233A (en) | 2017-08-22 | 2017-08-22 | A kind of image-recognizing method being applied in FOD foreign bodies detection radars |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107481233A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN108734679A (en) * | 2018-05-23 | 2018-11-02 | 西安电子科技大学 | A kind of computer vision system |
| CN109490301A (en) * | 2018-10-24 | 2019-03-19 | 深圳市锦润防务科技有限公司 | It is a kind of for monitor on floating platform adhere to analyte detection method, system and storage medium |
| CN109881437A (en) * | 2019-02-25 | 2019-06-14 | 珠海格力电器股份有限公司 | Inner cylinder, washing processing equipment and foreign matter detection method |
| CN110097533A (en) * | 2019-02-12 | 2019-08-06 | 哈尔滨新光光电科技股份有限公司 | A kind of method for accurate testing of hot spot outer dimension and position |
| CN110135296A (en) * | 2019-04-30 | 2019-08-16 | 上海交通大学 | Airfield runway FOD detection method based on convolutional neural networks |
| WO2019232831A1 (en) * | 2018-06-06 | 2019-12-12 | 平安科技（深圳）有限公司 | Method and device for recognizing foreign object debris at airport, computer apparatus, and storage medium |
| CN111568199A (en) * | 2020-02-28 | 2020-08-25 | 佛山市云米电器科技有限公司 | Method and system for identifying water receiving container and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105549110A (en) * | 2015-12-08 | 2016-05-04 | 北京无线电计量测试研究所 | Airport runway foreign object debris detection device and airport runway foreign object debris detection method |
| US20160221048A1 (en) * | 2015-02-04 | 2016-08-04 | The Boeing Company | System and method for high speed fod detection |
| CN106018409A (en) * | 2016-06-29 | 2016-10-12 | 哈尔滨工程大学 | Airfield runway foreign matter and crack recognizing system and detecting and recognizing method thereof |
-
2017
- 2017-08-22 CN CN201710724260.7A patent/CN107481233A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160221048A1 (en) * | 2015-02-04 | 2016-08-04 | The Boeing Company | System and method for high speed fod detection |
| CN105549110A (en) * | 2015-12-08 | 2016-05-04 | 北京无线电计量测试研究所 | Airport runway foreign object debris detection device and airport runway foreign object debris detection method |
| CN106018409A (en) * | 2016-06-29 | 2016-10-12 | 哈尔滨工程大学 | Airfield runway foreign matter and crack recognizing system and detecting and recognizing method thereof |
Non-Patent Citations (1)
| Title |
|---|
| 李煜: "机场跑道异物检测算法与系统设计研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN108734679A (en) * | 2018-05-23 | 2018-11-02 | 西安电子科技大学 | A kind of computer vision system |
| WO2019232831A1 (en) * | 2018-06-06 | 2019-12-12 | 平安科技（深圳）有限公司 | Method and device for recognizing foreign object debris at airport, computer apparatus, and storage medium |
| CN109490301A (en) * | 2018-10-24 | 2019-03-19 | 深圳市锦润防务科技有限公司 | It is a kind of for monitor on floating platform adhere to analyte detection method, system and storage medium |
| CN110097533A (en) * | 2019-02-12 | 2019-08-06 | 哈尔滨新光光电科技股份有限公司 | A kind of method for accurate testing of hot spot outer dimension and position |
| CN109881437A (en) * | 2019-02-25 | 2019-06-14 | 珠海格力电器股份有限公司 | Inner cylinder, washing processing equipment and foreign matter detection method |
| CN110135296A (en) * | 2019-04-30 | 2019-08-16 | 上海交通大学 | Airfield runway FOD detection method based on convolutional neural networks |
| CN111568199A (en) * | 2020-02-28 | 2020-08-25 | 佛山市云米电器科技有限公司 | Method and system for identifying water receiving container and storage medium |
| CN111568199B (en) * | 2020-02-28 | 2023-11-07 | 佛山市云米电器科技有限公司 | Water receiving container identification method, system and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107481233A (en) | A kind of image-recognizing method being applied in FOD foreign bodies detection radars | |
| CN109743879B (en) | Underground pipe gallery leakage detection method based on dynamic infrared thermography processing | |
| CN109583293B (en) | Aircraft target detection and identification method in satellite-borne SAR image | |
| Huang et al. | A new building extraction postprocessing framework for high-spatial-resolution remote-sensing imagery | |
| CN106856002B (en) | Unmanned aerial vehicle shooting image quality evaluation method | |
| CN110197231B (en) | Bird condition detection equipment and identification method based on visible light and infrared light image fusion | |
| CN110400267A (en) | A kind of preprocess method based on inspection image | |
| Li et al. | Road lane detection with gabor filters | |
| CN106709903B (en) | PM2.5 concentration prediction method based on image quality | |
| CN104361582A (en) | Method of detecting flood disaster changes through object-level high-resolution SAR (synthetic aperture radar) images | |
| CN110321855A (en) | A kind of greasy weather detection prior-warning device | |
| Fengping et al. | Road extraction using modified dark channel prior and neighborhood FCM in foggy aerial images | |
| CN110796677B (en) | Cirrus cloud false alarm source detection method based on multiband characteristics | |
| CN115984806B (en) | Dynamic detection system for road marking damage | |
| Jin et al. | Pavement crack detection fused HOG and watershed algorithm of range image | |
| Jiao et al. | Infrared dim small target detection method based on background prediction and high-order statistics | |
| Pratomo et al. | Parking detection system using background subtraction and HSV color segmentation | |
| Zhu et al. | A novel change detection method based on high-resolution SAR images for river course | |
| Dixit et al. | Comparison of effectiveness of dual tree complex wavelet transform and anisotropic diffusion in MCA for concrete crack detection | |
| Kaur et al. | An Efficient Method of Number Plate Extraction from Indian Vehicles Image | |
| Deng et al. | EMD based infrared image target detection method | |
| CN112508908B (en) | Method for detecting disconnection fault of sanding pipe joint of motor train unit based on image processing | |
| CN108665446A (en) | A kind of foreign body detection system for airfield runway and method with radar | |
| CN114140698A (en) | Water system information extraction algorithm based on FasterR-CNN | |
| Wang et al. | An automatic bridge detection technique for high resolution SAR images |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171215 |