CN105869156B - A kind of infrared small target detection method based on fuzzy distance - Google Patents
A kind of infrared small target detection method based on fuzzy distance Download PDFInfo
- Publication number
- CN105869156B CN105869156B CN201610177589.1A CN201610177589A CN105869156B CN 105869156 B CN105869156 B CN 105869156B CN 201610177589 A CN201610177589 A CN 201610177589A CN 105869156 B CN105869156 B CN 105869156B
- Authority
- CN
- China
- Prior art keywords
- fuzzy distance
- multiple dimensioned
- pixel
- distance
- infrared
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10048—Infrared image
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20004—Adaptive image processing
Landscapes
- Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
- Studio Devices (AREA)
- Image Analysis (AREA)
Abstract
The present invention is the infrared small target under effective detection of complex background, discloses a kind of infrared small target detection method based on fuzzy distance, is related to digital image processing techniques field.The present invention can cause local grain to occur to vary widely this feature first against the appearance of Small object, propose a kind of fuzzy distance concept, to convert the variation of local grain to the measurement of fuzzy distance；Secondly respective change this feature can occur with the change of image-forming range for the size of Small object, proposes a kind of multiple dimensioned fuzzy distance and multiple dimensioned fuzzy distance figure, a large amount of background clutters and noise jamming can be rejected；Then by interative computation, effectively inhibit residual background and noise, enhance target；Adaptive threshold is finally utilized to detect target, the detection method is simple and effective.
Description
Technical field
The present invention relates to digital image processing techniques field, specifically a kind of infrared small target detection based on fuzzy distance
Method.
Background technology
Infrared target detection technology is used widely in many civil fields, such as medicine infrared imaging, remote sensing and gloomy
Woods detection, early warning detection etc..Detection performance quality directly determines the EFFECTIVE RANGE of infrared system and the complexity of equipment
Degree.Remote infrared system is small so as to cause target size, intensity is weak because image-forming range is remote, and be easily submerged in very noisy and
In background clutter.In this case, effectively detect that the Small object of unknown position/speed/size/shape faces very disaster
Degree, thus this kind of technology receives and continues and universal concern.
Existing small target deteection technology can simply be divided into detection before tracking (Track before Detect, TBD) and
(Detect before Track, DBT) two classes are detected before tracking.TBD methods generally first search for all possible movement rail of target
Mark, and complete target energy and add up, to obtain the posterior probability of every movement locus, finally utilize the true mesh of threshold decision
Movement locus is marked, such as three-dimensional matched filtering, three-dimensional filtering.TBD methods be easy to establish relatively more complete theoretical model and
Processing method, but calculate complicated, hardware realization trouble, in Practical Project using it is less (C.Q.Gao, D.Y.Meng,
Y.Yang,Y.T.Wang,X.F.Zhou,A.G.Hauptmann,Infrared patch-image model for small
target detection in a single image,IEEE Transactions on Image Processing,22
(12):4996-5009,2013).DBT methods are general first to detect candidate target according to the gamma characteristic in short-term of single-frame images, then
False target is rejected according to the kinetic characteristic in short-term of target, to obtain the true movement locus of target.Compared with TBD methods,
DBT algorithms are simple, realized convenient for program modularity, thus play a significant role in real-time target detection field (H.Deng,
X.P.Sun,M.L.Liu,C.H.Ye,X.Zhou,Infrared small-target detection using
multiscale gray difference weighted image entropy,IEEE Transactions on
Aerospace and electronic Systems,52(1):60-72,2016).It is provided according to International Optical Engineering Society
Small object defines, and the size of Small object is usually no more than the 0.12% of entire image size, thus the appearance of target is to whole picture figure
Picture texture structure influence is smaller, but localized region texture structure is affected.Based on this feature, some describe local grain change
Change operator be suggested, can effectively detect infrared small target, such as multiple dimensioned intensity-weighted image entropy, sparse ring indicate, probability it is main at
Analysis, local contrast estimate (C.L.Philip, H.Li, Y.T.Wei, T.Xia, and Y.Y.Tang, A local
contrast method for small infrared target detection,IEEE Transactions on
Geoscience and Remote Sensing,51(1):574-581,2014).The application patented method is under the jurisdiction of the side DBT
Method.Compared with conventional DBT methods, the application patented method proposes a kind of fuzzy distance concept, can effectively portray target internal,
Inside background, the distance between target and background, to convert the local grain variation caused by the appearance of Small object to mould
The measurement of distance is pasted, realizes background inhibition, targets improvement, is conducive to the detection probability for improving target, reduces false-alarm probability.
Although infrared small target detection field has achieved many achievements, and has many TBD and DBT algorithms in engineering
Good realization has been obtained in.But for low signal-to-noise ratio infrared small target image under complex background, object detection system work
Cheng Yiran faces prodigious difficulty and complexity.Therefore, how to design that result is simple, infrared small target detection of strong robustness
Algorithm is the critical issue of target detection technique application study.
Invention content
The present invention be directed to above-mentioned technical problems existing for existing infrared small target detection method, provide a kind of based on mould
Paste the infrared small target detection method of distance.
A kind of infrared small target detection method based on fuzzy distance, includes the following steps：
Step 1, initialization relevant parameter：
Maximum iteration L is set, and wherein L is positive integer；It initializes iterations and indexes k=0；The maximum local window of setting
Mouth size m × n, wherein m and n are the positive odd number more than 1；
Step 2, the fuzzy distance for solving each pixels of infrared image I, include the following steps：
Step 2.1, the neighborhood space collection { Ω for obtaining each pixel (x, y) of Single Infrared Image Frame Il| l=1,
2 ..., s }, wherein s=min { 0.5 (m-1), 0.5 (n-1) }, ΩlSize be (2l+1) × (2l+1), pixel
The neighborhood space Ω of (x, y)lDefinition be Ωl=(p, q) | max (| p-x |, | q-y |)≤l, (p, q) is neighborhood space Ωl
Interior pixel；
Step 2.2, each neighborhood space Ω for calculating each pixel (x, y)lThe gray average D of interior pixell(x,y)：
Wherein, # ΩlIndicate neighborhood space ΩlThe number of interior pixel, I (a, b) indicate neighborhood space ΩlInterior pixel
Gray value at (a, b).
Step 2.3 calculates maximum neighborhood space Ω corresponding to each pixel (x, y)sIt is empty with other each neighborhoods
Between Ωi, i=1,2 ..., the fuzzy distance E between s-1i：
Wherein e is natural constant, DsIndicate maximum neighborhood space ΩsThe gray average of interior pixel, DiIndicate i-th of neighborhood
Space ΩiThe gray average of interior pixel；
Step 3 solves multiple dimensioned fuzzy distance figure：
Each pixel in infrared image I is traversed, obtains the multiple dimensioned fuzzy distance E (x, y) of each pixel, so
Afterwards the multiple dimensioned of infrared image I is obtained according to the multiple dimensioned fuzzy distance E (x, y) of each pixel and by method for normalizing
Fuzzy distance figure E；
Step 4, iteration stopping criterion judge:
Iterations index k adds 1, the relationship between iterations index k and maximum iteration L is judged, if k<L,
The multiple dimensioned fuzzy distance figure E that step 3 is obtained is as new infrared image I, return to step 2；If k >=L, stop iteration,
The multiple dimensioned fuzzy distance figure E that step 3 is obtained carries out step 5 as final filter result；
Step 5 solves adaptive threshold T：
To the final filter result obtained by step 4, i.e., multiple dimensioned fuzzy distance figure E solves adaptive threshold T,
And binaryzation is carried out to multiple dimensioned fuzzy distance figure E by adaptive threshold T, detect infrared small target.
The multiple dimensioned fuzzy distance of each pixel (x, y) of infrared image I is expressed as in step 3 as described above
E (x, y)=max { 0, E1,E2,…,Es-1}。
The determination method of adaptive threshold T is in step 5 as described above
T=α Emax+β·mt
Wherein, α and β is positive constant, mtFor the gray average of multiple dimensioned fuzzy distance figure E, EmaxFor it is multiple dimensioned it is fuzzy away from
Gray scale maximum value from figure E.
Compared with prior art, the present invention haing the following advantages：
1. the appearance for Small object can cause local grain to occur to vary widely this feature, the present invention proposes one kind
Fuzzy distance concept, can effectively portray target internal, inside background, the distance between target and background, to by local grain
Variation is converted into the measurement of fuzzy distance.
2. respective change this feature can occur with the change of image-forming range for the size for Small object, the present invention proposes
A kind of multiple dimensioned fuzzy distance measure reduces false-alarm probability to effectively improve the detection probability of target.
3. the present invention builds the multiple dimensioned fuzzy distance figure of infrared image under complex background first, it is miscellaneous a large amount of backgrounds to be rejected
Wave and noise jamming；Secondly by interative computation, effectively inhibit remaining background and noise, enhances target；Then it utilizes adaptive
It answers threshold value to detach target, simply and efficiently detects target.
Description of the drawings
Fig. 1 is the flow diagram of the present invention.
Fig. 2 is the multiple dimensioned fuzzy distance figure after successive ignition, wherein A figures are the Small object figure under original sky background
As (white rectangle frame indicates target region), B figures are the multiple dimensioned fuzzy distance figure after an iteration, and C figures are to change twice
Multiple dimensioned fuzzy distance figure after generation, D figures are the multiple dimensioned fuzzy distance figure after iteration three times, and E figures are more after four iteration
Scale fuzzy distance figure, F figures are the testing result using adaptive threshold.
Fig. 3 is the filter result schematic diagram of the single infrared small target image obtained using the present embodiment method.
A, B, C and D：Single infrared small target image (the white rectangle frame expression target being followed successively by under original different background
Region), wherein A figures are the Small object image under sky background, and B figures are the Small object image under clutter background, and C figures are
The underwater Small object image of marine background, D figures are the Small object image under surface feature background；
E, F, G and H：It is corresponding in turn to the filter result obtained using the present embodiment method in A, B, C and D.
Fig. 4 is the filter result schematic diagram of the two infrared small target images obtained using the present embodiment method.
A, B, C and D：Two infrared small target images (the white rectangle frame expression target being followed successively by under original different background
Region), wherein A figures are the Small object image under sky background, and B figures are the Small object image under sky background, and C figures are
Small object image under the water surface background of ocean, D figures are the Small object image under surface feature background；
E, F, G and H：It is corresponding in turn to the filter result obtained using the present embodiment method in A, B, C and D.
Fig. 5 is to use the filter result schematic diagram that existing method obtains for A, B, C and D in Fig. 3.
A1, B1, C1 and D1：Be corresponding in turn in Fig. 3 A, Fig. 3 B, Fig. 3 C and Fig. 3 D based on local contrast (Local
Contrast measure, LCM) method filter result；
A2, B2, C2 and D2：Be corresponding in turn in Fig. 3 A, Fig. 3 B, Fig. 3 C and Fig. 3 D based on Largest Mean filter (Maxmean
Filter, MME) method filter result；
A3, B3, C3 and D3：Be corresponding in turn in Fig. 3 A, Fig. 3 B, Fig. 3 C and Fig. 3 D based on max-medium filter
The filter result of (Maxmedian filter, MED) method；
A4, B4, C4 and D4：Be corresponding in turn in Fig. 3 A, Fig. 3 B, Fig. 3 C and Fig. 3 D based on top cap filter (Top-hat
Filter, THT) method filter result.
Fig. 6 is to use the filter result schematic diagram that existing method obtains for A, B, C and D in Fig. 4.
A1, B1, C1 and D1：Be corresponding in turn in Fig. 4 A, Fig. 4 B, Fig. 4 C and Fig. 4 D based on local contrast (Local
Contrast measure, LCM) method filter result；
A2, B2, C2 and D2：Be corresponding in turn in Fig. 4 A, Fig. 4 B, Fig. 4 C and Fig. 4 D based on Largest Mean filter (Maxmean
Filter, MME) method filter result；
A3, B3, C3 and D3：Be corresponding in turn in Fig. 4 A, Fig. 4 B, Fig. 4 C and Fig. 4 D based on max-medium filter
The filter result of (Maxmedian filter, MED) method；
A4, B4, C4 and D4：Be corresponding in turn in Fig. 4 A, Fig. 4 B, Fig. 4 C and Fig. 4 D based on top cap filter (Top-hat
Filter, THT) method filter result.
Specific implementation mode
Below with reference to the embodiments and with reference to the accompanying drawing the technical solutions of the present invention will be further described.
Embodiment：
Fig. 1 is the structural schematic block diagram of embodiment of the present invention, and main includes-image input：Input the infrared small mesh of single frames
Logo image；Fuzzy distance solves：Solve the fuzzy distance between input picture current region and neighborhood collection；Multiple dimensioned fuzzy distance
Figure solves：The multiple dimensioned fuzzy distance figure for solving input picture, is portrayed that may be present caused by the variation of image-forming range
Target size changes this characteristic；Iteration stopping judges：Judge the relationship between iterations index and maximum iteration, weight
Multiple iteration, obtains final filter result；Threshold value solves：Solve the segmentation threshold of final filter result；Binaryzation：Pass through threshold value point
From target, target centroid position is obtained.
Specially：
Step 1, relevant parameter is initialized：
Be arranged maximum iteration L, wherein L be positive integer, generally 2,3 or 4；It initializes iterations and indexes k=0；
It is the positive odd number more than 1 that maximum local window size m × n, wherein m and n, which is arranged, the value of general m and n be disposed as 7,9 or
11。
Step 2, the fuzzy distance of each pixels of infrared image I is solved：
Infrared small target image under complex background is generally made of target, complex background and noise three parts.Degree of passing through
Measure target internal, inside background, the distance between target and background, to by the local grain caused by the appearance of Small object
Variation is converted into the measurement of fuzzy distance, realizes background inhibition and targets improvement.
The solution procedure of the fuzzy distance of each pixels of infrared image I is as follows：
(1) the neighborhood space collection { Ω of each pixel (x, y) of Single Infrared Image Frame I is obtainedl| l=1,2 ..., s },
Middle s=min { 0.5 (m-1), 0.5 (n-1) }, ΩlSize be (2l+1) × (2l+1), the neighborhood of pixel (x, y) is empty
Between ΩlDefinition be Ωl=(p, q) | max (| p-x |, | q-y |)≤l, (p, q) is neighborhood space ΩlInterior pixel.
(2) each neighborhood space Ω of each pixel (x, y) is calculatedlThe gray average D of interior pixell(x,y)：
Wherein, # ΩlIndicate neighborhood space ΩlThe number of interior pixel, I (a, b) indicate neighborhood space ΩlInterior pixel
Gray value at (a, b).
(3) the maximum neighborhood space Ω corresponding to each pixel (x, y) is calculatedsWith other each neighborhood space Ωi,i
=1,2 ..., the fuzzy distance E between s-1i：
Wherein e is natural constant, DsIndicate maximum neighborhood space ΩsThe gray average of interior pixel, DiIndicate i-th of neighborhood
Space ΩiThe gray average of interior pixel.
Step 3, multiple dimensioned fuzzy distance figure is solved：
Although lacking the prioris such as infrared small target size, size, with the change of image-forming range, the ruler of target
The features such as very little, size can occur to change to a certain extent.It is portrayed using multiple dimensioned fuzzy distance that may be present because of image-forming range
Variation caused by target size change this property.
The multiple dimensioned fuzzy distance of each pixel (x, y) of infrared image I is expressed as
E (x, y)=max { 0, E1,E2,...,Es-1} (3)
Wherein, Ei, i=1,2 ..., s-1 indicates a series of fuzzy distances of pixel (x, y).
Each pixel in infrared image I is traversed, the multiple dimensioned fuzzy distance of each pixel is obtained, then passes through
Method for normalizing obtains the multiple dimensioned fuzzy distance figure E of infrared image I (as shown in the B of Fig. 2).It can be seen that from the B of Fig. 2
The homogeneous sky background and cloud layer interior background of infrared image I is inhibited, and target is enhanced.
Step 4, iteration stopping criterion judges:
Iterations index k adds 1, the relationship between iterations index k and maximum iteration L is judged, if k<L, will
The multiple dimensioned fuzzy distance figure E that step 3 is obtained is as new infrared image I, return to step 2；If k >=L, stop iteration, it will
The multiple dimensioned fuzzy distance figure that step 3 is obtained carries out step 5 as final filter result.
The complex background boundary of infrared image has thermal imaging feature similar with target, can be with by the way that iteration is repeated several times
Eliminate the influence (as shown in Figure 2) of residual background and noise.The B of Fig. 2 indicates the multiple dimensioned fuzzy distance figure after an iteration.From
As can be seen that the homogeneous background in the B of Fig. 2 is inhibited well (inside homogeneous sky and homogeneous cloud layer) in the B of Fig. 2, but
Remain more cloud layer edge.It is remaining that the multiple dimensioned fuzzy distance figure (as shown in the C of Fig. 2) of the B of Fig. 2 can remove the overwhelming majority
Cloud layer edge, and the multiple dimensioned fuzzy distance figure (as shown in the D of Fig. 2) of the C of Fig. 2 can remove remaining cloud layer edge so that it is multiple
Miscellaneous cloud layer boundary is further inhibited, and target is further enhanced from.The multiple dimensioned fuzzy distance figure of the D of Fig. 2 is (as schemed
Shown in 2 E) it is little with the D differences of Fig. 2, illustrate to be obtained with comparatively ideal filtering knot using suitable limited iterations
Fruit.
Step 5, adaptive threshold T is solved：
Adaptive threshold T is solved to the final filter result (i.e. multiple dimensioned fuzzy distance figure E) obtained by step 4,
And binaryzation is carried out to multiple dimensioned fuzzy distance figure E by adaptive threshold T, detecting infrared small target, (binaryzation result is such as
Shown in the F of Fig. 2).The determination method of adaptive threshold T is
T=α Emax+β·mt (4)
Wherein, α and β is positive constant, mtFor the gray average of multiple dimensioned fuzzy distance figure E, EmaxFor multiple dimensioned fuzzy distance figure
The gray scale maximum value of E.
Filter result using different infrared small target detection methods is as shown in Figure 5 and Figure 6.Compare Fig. 3, Fig. 4, Fig. 5 and
Fig. 6, the filtering performance that the present embodiment method obtains are best, wherein are based on local contrast (Local contrast
Measure, LCM) method comes from document C.L.Philip, H.Li, Y.T.Wei, T.Xia, and Y.Y.Tang, A local
contrast method for small infrared target detection,IEEE Transactions on
Geoscience and Remote Sensing,51(1):574-581,2014, LCM methods first pass through local contrast measure
Dissimilarity between current region and neighborhood, then detaches target by threshold value；(Maxmean is filtered based on Largest Mean
Filter, MME) or max-medium filter (Maxmedian filter, MED) method come from document S.Deshpande,
M.Er,and R.Venkateswarlu,Maxmean and Maxmedian filters for detection of
small-targets,Proceeding of SPIE,1999,3809:74-83, MME/MED method be first pass through Largest Mean/
Median filter wiping out background noise jamming detaches target then according to the statistical property threshold value of image；It is filtered based on top cap
Wave (Top-hat filter, THT) method comes from document X.Z.Bai and F.G.Zhou, Analysis of new top-
hat transformation and the application for infrared dim small target
detection,Pattern Recognition,2010,43(6):2145-2156, THT method first pass through top-cap operator and inhibit
Then background and noise detach target using threshold value from filtered image.LCM, MME, MED, THT are under the jurisdiction of DBT methods.
Using Background suppression factor (Background suppression factor, BSF) objective evaluation infrared small target
The filtering performance of detection method.The definition of BSF is：
BSF=σI/σO (5)
Wherein, σIIndicate the gray standard deviation of filtered image, σOIndicate the gray standard deviation of filter wavefront image.Using
The BSF numerical value that LCM, MME, MED, THT and the present embodiment method are obtained is shown in Table 1.As it can be seen from table 1 the present embodiment method
Highest BSF values are obtained, illustrate that the present embodiment method can effectively inhibit the complex background and noise of infrared small target image, with
Fig. 5 is consistent with the conclusion that Fig. 6 is obtained.
Table 1 is compared using the Background suppression factor (BSF) of different infrared small target detection methods
| Image | LCM methods | MME methods | MED methods | THT methods | The present embodiment method |
| Fig. 3 A | 0.8134 | 0.5638 | 0.5826 | 1.0157 | 5.8339 |
| Fig. 3 B | 0.9348 | 0.7368 | 0.7396 | 1.8542 | 8.2408 |
| Fig. 3 C | 0.5686 | 0.3288 | 0.3521 | 0.7993 | 6.0785 |
| Fig. 3 D | 0.4272 | 0.3800 | 0.3843 | 1.6548 | 6.3409 |
| Fig. 4 A | 0.4179 | 0.2640 | 0.2476 | 1.6413 | 1.7025 |
| Fig. 4 B | 0.1698 | 0.1259 | 0.1232 | 0.3681 | 0.5156 |
| Fig. 4 C | 0.2227 | 0.2171 | 0.2164 | 1.0775 | 4.7691 |
| Fig. 4 D | 0.5685 | 0.5246 | 0.5227 | 1.6320 | 7.5393 |
Specific embodiment described herein is only an example for the spirit of the invention.Technology belonging to the present invention is led
The technical staff in domain can make various modifications or additions to the described embodiments or replace by a similar method
In generation, however, it does not deviate from the spirit of the invention or beyond the scope of the appended claims.
Claims (2)
1. a kind of infrared small target detection method based on fuzzy distance, which is characterized in that include the following steps：
Step 1, initialization relevant parameter：
Maximum iteration L is set, and wherein L is positive integer；It initializes iterations and indexes k=0；The maximum local window of setting is big
Small m × n, wherein m and n are the positive odd number more than 1；
Step 2, the fuzzy distance for solving each pixels of infrared image I, include the following steps：
Step 2.1, the neighborhood space collection { Ω for obtaining each pixel (x, y) of Single Infrared Image Frame Il| l=1,2 ..., s },
Wherein s=min { 0.5 (m-1), 0.5 (n-1) }, ΩlSize be (2l+1) × (2l+1), the neighborhood of pixel (x, y)
Space ΩlDefinition be Ωl=(p, q) | max (| p-x |, | q-y |)≤l, (p, q) is neighborhood space ΩlInterior pixel；
Step 2.2, each neighborhood space Ω for calculating each pixel (x, y)lThe gray average D of interior pixell(x,y)：
Wherein, # ΩlIndicate neighborhood space ΩlThe number of interior pixel, I (a, b) indicate neighborhood space ΩlInterior pixel (a, b)
The gray value at place,
Step 2.3 calculates maximum neighborhood space Ω corresponding to each pixel (x, y)sWith other each neighborhood space Ωi,
I=1,2 ..., the fuzzy distance E between s-1i：
Wherein e is natural constant, DsIndicate maximum neighborhood space ΩsThe gray average of interior pixel, DiIndicate i-th of neighborhood space
ΩiThe gray average of interior pixel；
Step 3 solves multiple dimensioned fuzzy distance figure：
Each pixel in infrared image I is traversed, obtains the multiple dimensioned fuzzy distance E (x, y) of each pixel, then root
According to each pixel multiple dimensioned fuzzy distance E (x, y) and pass through method for normalizing and obtain the multiple dimensioned fuzzy of infrared image I
Distance map E；
Step 4, iteration stopping criterion judge:
Iterations index k adds 1, the relationship between iterations index k and maximum iteration L is judged, if k<L, step 3
The multiple dimensioned fuzzy distance figure E obtained is as new infrared image I, return to step 2；If k >=L, stop iteration, step 3
The multiple dimensioned fuzzy distance figure E obtained carries out step 5 as final filter result；
Step 5 solves adaptive threshold T：
To the final filter result obtained by step 4, i.e., multiple dimensioned fuzzy distance figure E solves adaptive threshold T, and leads to
It crosses adaptive threshold T and binaryzation is carried out to multiple dimensioned fuzzy distance figure E, detect infrared small target,
The multiple dimensioned fuzzy distance of each pixel (x, y) of infrared image I is expressed as in step 3
E (x, y)=max { 0, E1,E2,...,Es-1}。
2. a kind of infrared small target detection method based on fuzzy distance according to claim 1, which is characterized in that described
Step 5 in the determination method of adaptive threshold T be
T=α Emax+β·mt
Wherein, α and β is positive constant, mtFor the gray average of multiple dimensioned fuzzy distance figure E, EmaxFor multiple dimensioned fuzzy distance figure E
Gray scale maximum value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610177589.1A CN105869156B (en) | 2016-03-25 | 2016-03-25 | A kind of infrared small target detection method based on fuzzy distance |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610177589.1A CN105869156B (en) | 2016-03-25 | 2016-03-25 | A kind of infrared small target detection method based on fuzzy distance |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105869156A CN105869156A (en) | 2016-08-17 |
| CN105869156B true CN105869156B (en) | 2018-07-17 |
Family
ID=56625266
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610177589.1A Active CN105869156B (en) | 2016-03-25 | 2016-03-25 | A kind of infrared small target detection method based on fuzzy distance |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105869156B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107590496A (en) * | 2017-09-18 | 2018-01-16 | 南昌航空大学 | The association detection method of infrared small target under complex background |
| CN108198198A (en) * | 2017-12-22 | 2018-06-22 | 湖南源信光电科技股份有限公司 | Single frames infrared small target detection method based on wavelet transformation and Steerable filter |
| CN109886980B (en) * | 2019-03-04 | 2023-04-14 | 电子科技大学 | Infrared image cirrus cloud detection method based on neighborhood intensity texture coding |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104463911A (en) * | 2014-12-09 | 2015-03-25 | 上海新跃仪表厂 | Small infrared moving target detection method based on complicated background estimation |
| CN104657945A (en) * | 2015-01-29 | 2015-05-27 | 南昌航空大学 | Infrared small target detection method for multi-scale spatio-temporal union filtering under complex background |
| CN104899866A (en) * | 2015-05-05 | 2015-09-09 | 河南三联网络技术有限公司 | Intelligent infrared small target detection method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7362880B2 (en) * | 2005-03-31 | 2008-04-22 | Lockheed Martin Corporation | Unresolved target detection improvement by use of multiple matched filters approach at different spatial phases |
-
2016
- 2016-03-25 CN CN201610177589.1A patent/CN105869156B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104463911A (en) * | 2014-12-09 | 2015-03-25 | 上海新跃仪表厂 | Small infrared moving target detection method based on complicated background estimation |
| CN104657945A (en) * | 2015-01-29 | 2015-05-27 | 南昌航空大学 | Infrared small target detection method for multi-scale spatio-temporal union filtering under complex background |
| CN104899866A (en) * | 2015-05-05 | 2015-09-09 | 河南三联网络技术有限公司 | Intelligent infrared small target detection method |
Non-Patent Citations (3)
| Title |
|---|
| An improved algorithm for facet-based infrared small target detection;KEJIA YI 等;《JOURNAL OF OPTOELECTRONICS AND ADVANCED MATERIALS》;20120430;第14卷(第3-4期);298-303 * |
| 基于多尺度局部方差的海面红外舰船检测;邹常文 等;《激光与红外》;20110620;第41卷(第6期);697-699 * |
| 基于邻域对比的红外林火小目标检测;张红民 等;《重庆理工大学学报（自然科学）》;20150115;第29卷(第1期);107-110、130 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105869156A (en) | 2016-08-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103942803B (en) | SAR (Synthetic Aperture Radar) image based automatic water area detection method | |
| CN102609701B (en) | Remote sensing detection method based on optimal scale for high-resolution SAR (synthetic aperture radar) | |
| CN105528619B (en) | SAR remote sensing image variation detection method based on wavelet transformation and SVM | |
| CN106170819A (en) | A kind of diameter radar image Ship Target method for quick | |
| CN104268877B (en) | A kind of infrared image sea horizon self-adapting detecting method | |
| CN106886747B (en) | It is a kind of based on extension wavelet transformation complex background under Ship Detection | |
| CN104834915B (en) | A kind of small infrared target detection method under complicated skies background | |
| CN105869156B (en) | A kind of infrared small target detection method based on fuzzy distance | |
| CN107358161B (en) | Coastline extraction method and coastline extraction system based on remote sensing image classification | |
| CN114372960B (en) | Airplane ground ice accumulation detection method, system, terminal and medium based on image understanding | |
| CN115983141B (en) | Deep learning-based method, medium and system for inverting wave height of ocean waves | |
| CN106157264A (en) | Large area image uneven illumination bearing calibration based on empirical mode decomposition | |
| Chen et al. | Marine target detection based on Marine-Faster R-CNN for navigation radar plane position indicator images | |
| CN106803078A (en) | A kind of SAR image Ship Target dividing method | |
| Zheng et al. | Adaptive edge detection algorithm based on grey entropy theory and textural features | |
| CN107464255A (en) | A kind of ship target detection method based on information content Yu multiple dimensioned abnormality detection | |
| Li et al. | An infrared dim and small target detection method based on fractional differential | |
| CN113689449B (en) | Inversion method and system for characteristic parameters of mesoscale vortices | |
| CN111951299B (en) | Infrared aerial target detection method | |
| Fu et al. | Recognition of bridge over water in high-resolution remote sensing images | |
| CN114219996A (en) | SAR image berthing ship target detection method | |
| Zhang et al. | Ship target segmentation and detection in complex optical remote sensing image based on component tree characteristics discrimination | |
| CN113516187A (en) | Infrared weak and small target detection algorithm adopting local characteristic contrast | |
| Liu et al. | Adaptive ideal image reconstruction for bacteria colony detection | |
| Jeong et al. | Automatic image contrast enhancement for small ship detection and inspection using RADARSAT-2 synthetic aperture radar data |
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 |