CN109816641A - Weighted local entropy infrared small target detection method based on Multiscale Morphological Fusion - Google Patents
Weighted local entropy infrared small target detection method based on Multiscale Morphological Fusion Download PDFInfo
- Publication number
- CN109816641A CN109816641A CN201910017099.9A CN201910017099A CN109816641A CN 109816641 A CN109816641 A CN 109816641A CN 201910017099 A CN201910017099 A CN 201910017099A CN 109816641 A CN109816641 A CN 109816641A
- Authority
- CN
- China
- Prior art keywords
- image
- img
- wths
- hat
- entropy
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The present invention proposes a kind of weighted local entropy infrared small target detection method based on Multiscale Morphological image co-registration, handles firstly, being converted infrared image to gray scale domain；Secondly, Multiscale Morphological Top-Hat image dividing processing is carried out to infrared image, image difference is sought on the basis of the Top-Hat of adjacent scale, obtains minimum difference figure, minimum difference figure is merged with the minimum mean figure of the image converted through Top-Hat again, obtains the image after background inhibits；Then, by calculating the local entropy of initial pictures, local entropy hum pattern is obtained；Later, image and local entropy hum pattern after inhibiting to background carry out dot product, and normalize, and obtain the notable figure of infrared small target；Finally, image after being handled, the region that two-value therein turns to 1 is exactly infrared small target using Threshold sementation to the filtering of infrared small target notable figure and binaryzation.The present invention is suitable for infrared small target detection field, can effectively improve the accuracy rate of infrared small target detection, and effectively reduce false alarm rate.
Description
Technical field
The invention belongs to digital image processing techniques field, specially a kind of weighting office based on Multiscale Morphological Fusion
Portion's entropy infrared small target detection method.
Background technique
Small IR target detection has become defense military, the important branch of civil field, due to infrared acquisition skill
Art has many advantages, such as that round-the-clock, detection range is remote, hiding is good, infrared remote target following, infrared terminal guidance and
It is widely used on the fields such as border national defence, health care.But since the imaging sensor of infrared image is by device sheet
The influence of the inherent characteristics of the factors such as body temperature, material and infrared small target image itself, in single-frame images, so that red
Outer Small object imaging Small Target pixel accounting is small, signal is weak, vulnerable to interference, and eventually leading to target detection and tracking becomes abnormal
Suffering.Therefore, infrared small target detection becomes the hot and difficult issue in digital image processing field research.
With the fast development of information technology field, the scale of image data becomes increasing, in face of so it is huge and
How numerous data fast and accurately complete image analysis, have been transformed into an important branch of current research.
Since Laurent Itti in 1998 proposes a kind of vision significance detection method, vision significance detection technique is got over
Carry out the concern of more scholars.Vision significance detection is to simulate primate in the field computer vision (Computer-Vision)
The vision noticing mechanism of animal and the project occurred.In recent years, scholars are based on the mechanism and propose many related algorithms
And apply to infrared small target detection, tracking field.View-based access control model attention mechanism extracts the significant properties of infrared small target, is used in combination
In infrared small target detection, good testing result can be obtained.Chen et al. proposes a kind of based on human eye vision comparison
The local contrast of mechanism (Human Visual System, HVS) measures (Local Contrast Method, LCM) algorithm.
It is significant to protrude Small object according to the difference of local contrast (Local Contrast, LC) feature of target and background for LCM algorithm
Property, inhibit background clutter, and obtain the notable figure of infrared small target image.Han et al. proposes a kind of improved LCM algorithm, claims
For improved local contrast algorithm (Improve Local Contrast Method, ILCM), which is again based on HVS
Algorithm.LCM algorithm is by improving, and calculating speed is very fast, and false alarm rate also reduces, but the detection under complex background is still one
Item problem.
Summary of the invention
For the deficiency in infrared image small target deteection field of the above prior art under complex background, the present invention is proposed
A kind of weighted local entropy (Weighted Local Entropy of Multi- based on Multiscale Morphological image co-registration
Scale Morphological Image Fusion, WLEMMF) Method of Target Detection in Infrared, this detection method is suitable for
Infrared small target detection field can effectively improve the accuracy rate of infrared small target detection, and effectively reduce false alarm rate.
Realizing algorithm thinking of the invention is: handling firstly, being converted infrared image to gray scale domain；Secondly,
According to the WLEMMF algorithm of proposition, Multiscale Morphological Top-Hat image dividing processing is carried out to infrared image, in adjacent scale
Top-Hat on the basis of seek image difference, obtain minimum difference figure, then by minimum difference figure and the figure that converts through Top-Hat
The minimum mean figure of picture merges, and obtains the image after background inhibits；Then, it by calculating the local entropy of initial pictures, obtains
Local entropy hum pattern；Later, image and local entropy hum pattern after inhibiting to background carry out dot product, and normalize, and obtain infrared
The notable figure of Small object；Finally, using Threshold sementation to the filtering of infrared small target notable figure and binaryzation, after obtaining processing
Image, spot zone (two-value turns to 1 region) therein is exactly infrared small target.
Based on the above principles, the technical solution of the present invention is as follows:
A kind of weighted local entropy infrared small target detection method based on Multiscale Morphological Fusion, feature exist
In: the following steps are included:
Step 1: one infrared image Img to be detected of input, image size are m*n；
Step 2: the image Img of input being converted into gray level image, obtains image Imgin；
Step 3: background inhibition is carried out using following steps:
Step 3.1: I morphology pre-processing structure unit m of settingi, 1≤i≤I, wherein I takes the integer greater than zero, and
Morphology pre-processing structure unit scale increases with i and is increased；
Step 3.2: selection morphology pre-processing structure unit miTo image ImginTop-Hat morphological images are carried out to locate in advance
Reason, obtains pretreated image, is denoted as WTHSi；
Step 3.3: repeat step 3.2, until all settings morphology pre-processing structure unit all with image Imgin
The pretreatment of Top-Hat morphological images is completed, multiple dimensioned Top-Hat pretreatment image collection WTHS is obtained；
Step 3.4: to the gray scale of adjacent two images corresponding pixel points in multiple dimensioned Top-Hat pretreatment image collection WTHS
It is poor that value is made, and obtains disparity map:
WTHSTop_Hat(i+1)=WTHSi+1-WTHSi, 1≤i≤I-1
Step 3.5: repeating step 3.4, obtain adjacent two width of I-1 group in multiple dimensioned Top-Hat pretreatment image collection WTHS
The difference atlas WTHS of disparity map composition between imageTop_Hat；
Step 3.6: to difference atlas WTHSTop_HatThe fusion of element gray value minimum is carried out, blending image is obtained
WTHSTop_Hat_min；The gray average for calculating each image in multiple dimensioned Top-Hat pretreatment image collection WTHS, take it is therein most
Small mean value figure WTHSmean(min)；
Step 3.7: by image WTHSTop_Hat_minWith minimum mean figure WTHSmean(min)Middle corresponding pixel points gray value phase
Add, obtains the fusion figure Img after background inhibitsWTH；
Step 4: calculating image ImginImage local gray scale comentropy:
Step 4.1: choosing radius is the comentropy calculation window of M to image ImginFrom top to bottom, from left to right carry out office
Portion's grayscale information entropy operation: the center of comentropy calculation window is from image ImginThe position (M+1, M+1) start to slide, count simultaneously
The local message entropy under the window is calculated, image Img is to the last slided intoinThe position (m-M, n-M) until, image ImginSide
Edge comentropy is filled with 0, and obtaining local gray level comentropy seal is Imgentropy；
Step 4.2: by local gray level information entropy diagram ImgentropyImg is schemed with mergingWTHThe pixel value phase of corresponding pixel points
Multiply, and obtains weighting local gray level information entropy diagram Img after normalizingL_entropyAs infrared small target notable figure；
Step 5: calculate grayscale information entropy adaptive threshold:
Step 5.1: calculating infrared small target notable figure ImgL_entropyPixel value mean value Mean and pixel value variance Var,
And find out the maximum value H of wherein pixel valuemax；
Step 5.2: calculating threshold value T=α (Hmax-Mean)+βMean2+ ε Var, wherein α, β, ε are the normal number of setting；
Step 6: the threshold value T obtained using step 5 is to infrared small target notable figure ImgL_entropyBinary conversion treatment is carried out,
Pixel value is set to 0 lower than T in figure, is set to 1 not less than T；Obtain the binarization result of infrared small target notable figure
ImgL_entropy_BIt is distributed detection figure as final infrared small target, it is exactly infrared small target that wherein two-value, which turns to 1 region,.
Beneficial effect
The present invention compared with prior art, has the advantage that
1) it proposes a kind of new algorithm WLEMMF, image is handled with Multiscale Morphological Top-Hat, is schemed to obtained by different scale
As calculating image difference, and under each scale on the basis of disparity map, each pixel minimum difference figure is merged to obtain, is enhanced
Type morphology pre-processes background and inhibits image, the new weight as image grayscale comentropy；
2) suitable scale window radius is selected, the calculating of local gray level comentropy is carried out, enhanced Morphological scale-space is calculated
Method is merged with local gray level information entropy diagram, is inhibited to extended background, and effectively retains Small object information, obtains small mesh
Mark notable figure；
3) final notable figure is to weight, and adaptive image threshold algorithm is introduced, by the information of notable figure with two
The form of value highlights, and obtains final infrared small target notable figure.
Additional aspect and advantage of the invention will be set forth in part in the description, and will partially become from the following description
Obviously, or practice through the invention is recognized.
Detailed description of the invention
Above-mentioned and/or additional aspect of the invention and advantage will become from the description of the embodiment in conjunction with the following figures
Obviously and it is readily appreciated that, in which:
Fig. 1 is the fundamental block diagram of small IR targets detection.
Fig. 2 is the result example 1 of small IR targets detection: wherein in the first row, first is classified as infrared original input picture；
Second is classified as the Saliency maps picture by obtaining after this algorithm process；Third is classified as adaptive algorithm testing result.Second row
In, first is classified as the space diagram of infrared original input picture；Second is classified as the Saliency maps by obtaining after this algorithm process
The space diagram of picture；Third is classified as the space diagram of adaptive algorithm testing result.
Fig. 3 is the result example 2 of small IR targets detection；
Fig. 4 is the result example 3 of small IR targets detection；
Fig. 5 is the result example 4 of small IR targets detection；
Fig. 6 is the result example 5 of small IR targets detection；
Fig. 7 is the result example 6 of small IR targets detection.
Specific embodiment
The embodiment of the present invention is described below in detail, the embodiment is exemplary, it is intended to it is used to explain the present invention, and
It is not considered as limiting the invention.
Referring to Fig.1, the step of present invention realizes is as follows:
Step 1: one infrared image Img to be detected of input, image size are m*n.
Step 2: the image Img of input being converted into gray level image, obtains image Imgin。
Step 3: background inhibits:
Based on being expanded in morphology and corroding basic operation, if f (i, j) is initial gray level image, m (x, y)
For the structural unit for expanding and corroding.Expansive working is carried out to initial gray level image with structural unit m, is rememberedDefinition
Are as follows:
In formula, Df、DmIt is the domain of f (i, j) and m (x, y) respectively.Corrosion note f Θ m, is defined as:
(f Θ m) (i, j)=min f (i-x, j-y)-m (x, y) | (i-x, j-y) ∈ Df, (x, y) ∈ Dm}
Expansion and corrosion are used as morphologic basic operation, and expansion can be used as local maximum filtering operation, and corrode
It can be used as local minimum filtering operation.Based on this, opening operation f ο m and closing operation fm are derived,
Is defined as:
Opening operation is first to be corroded to initial pictures with structural unit m (x, y), then carry out expansion behaviour to result images
Make, doing so can inhibit to encourage point to a certain extent, and closed operation is first swollen to initial pictures progress with structural unit m (x, y)
It is swollen, etching operation is being carried out to result images.
By initial pictures subtract opening operation as a result, this Operation Definition is that top cap converts (Top-Hat), which can be with
The point of some edges of regions in image is obtained, this provides a kind of thinking in grayscale image, finding Small object.Definition
Are as follows:
G=f- (f ο m)
Top-Hat under single scale can not effectively reject the isolated noise in image.
This algorithm improves traditional morphology Top-Hat algorithm, forms a kind of enhanced morphology Top-Hat.
This algorithm needle initial pictures carry out the transformation of the Top-Hat under a variety of scales, and by institute under different scale (scale is from small to large)
The pre-filtered image WTHS obtainedi, wherein i is scale index, and the Top-Hat under adjacent scale is merged, and obtains image difference
Different atlas WTHSTop_Hat.It to obtained disparity map, is merged, fusion obtains the smallest disparity map WTHSTop_Hat_min.And
The wherein the smallest figure of mean value and WTHS are found under image difference mapTop_Hat_minDoing image adds operation to be merged, after fusion
Scheme ImgWTH。
According to International Optical Engineering Society (Society of Photo-Optical Instrumentation
Engineers, SPIE) regulation, a width size is the infrared image of m*n, and infrared small target pixel number is not super to occupy place
The 0.12% of pixel in image.I morphology pre-processing structure unit is set, wherein I takes the integer greater than zero.In computer
M in fieldiIndicate that size is (2k+1)2Square area, selection set in max architecture mIUnit scale (2k+1)2≤
0.0012*m*n。
Wherein 1≤i≤I, in order to reduce omission factor, miConditions above need to be met, therefore scale is selected and can be schemed according to input
As the priori knowledge of size is set.
Step 3.1: I morphology pre-processing structure unit m of settingi, 1≤i≤I, wherein I takes the integer greater than zero, and
Morphology pre-processing structure unit scale increases with i and is increased；
Step 3.2: selection morphology pre-processing structure unit miTo image ImginTop-Hat morphological images are carried out to locate in advance
Reason, obtains pretreated image, is denoted as WTHSi；WTHSi=Imgin-(Imginοmi), ImginFor the grayscale image of input picture,
' ο ' is the opening operation in morphology；
Step 3.3: repeat step 3.2, until all settings morphology pre-processing structure unit all with image Imgin
The pretreatment of Top-Hat morphological images is completed, multiple dimensioned Top-Hat pretreatment image collection WTHS is obtained；
Step 3.4: to the gray scale of adjacent two images corresponding pixel points in multiple dimensioned Top-Hat pretreatment image collection WTHS
It is poor that value is made, and obtains disparity map:
WTHSTop-Hat(i+1)=WTHSi+1-WTHSi, 1≤i≤I-1
Step 3.5: repeating step 3.4, obtain adjacent two width of I-1 group in multiple dimensioned Top-Hat pretreatment image collection WTHS
The difference atlas WTHS of disparity map composition between imageTop_Hat；
Step 3.6: to difference atlas WTHSTop_HatThe fusion of element gray value minimum is carried out, i.e. extraction disparity map is concentrated
Each pixel gray value minimum value of the correspondence of each figure is fused into a secondary new image, obtains blending image WTHSTop_Hat_min；Meter
Calculate the gray average of each image in multiple dimensioned Top-Hat pretreatment image collection WTHS, i.e. pre-filtering figure under i-th of scale
WTHSiMean value WTHSmean(i), minimum mean figure therein is selected, WTHS is denoted asmean(min)；
Step 3.7: by image WTHSTop_Hat_minWith minimum mean figure WTHSmean(min)Middle corresponding pixel points gray value phase
Add, obtains the fusion figure Img after background inhibitsWTH；
ImgWTH (j, k)=WTHSTop_Hat_min (j, k)+WTHSMean (min) (j, k)
Wherein subscript (j, k) is the coordinate of respective pixel.
Step 4: calculating image ImginImage local gray scale comentropy:
In a frame infrared image, infrared small target can regard that one exists in the background as, and not compatible with background
Singular point, this characteristic helps to come out Small object from complicated background separation, and is detected.But when background is complex
When, when noise jamming is stronger, easily target information is flooded, this results in Small object to be difficult to separate from background, and detection can not
Determine real goal.And image grayscale comentropy, it is a kind of statistics of spuious feature of information, reflects the average information of image
Amount indicates the aggregation characteristic of image grayscale distribution.And select suitable window function, it will be able to avoid Small object quilt to a greater extent
Missing inspection.
Grayscale information entropy in image, for the difference of window function size, the image grayscale distributed intelligence of extraction slightly has difference,
Window is smaller, can preferably extract the Small object information in grayscale information.But often window is too small, and information is caused excessively to mention
It takes, some isolated noise spots is also effectively counted, the complexity of calculating is not only increased, and reduce Small object
The robustness of detection, it is therefore desirable to select suitable entropy window size M, the present invention is arranged windows radius and is set as 2.Calculate the field
Grayscale information entropy, and replace the pixel value of window center position that the image can be obtained after traversing whole image with this entropy
Local entropy hum pattern.
Step 4.1: selection radius is M, and window size is (2M+1)2Comentropy calculation window to image ImginOn to
Under, from left to right carry out r local gray level comentropy operation；
0≤r≤(m-2M)*(n-2M)
If this window space is Ωθ, the center of comentropy calculation window is from image ImginThe position (M+1, M+1) start to slide
It is dynamic, while the local message entropy under the window is calculated, to the last slide into image ImginThe position (m-M, n-M) until, whenever
When window center one pixel of movement, all statistical information needs are recalculated；Image ImginMarginal information entropy be filled with
0, obtaining local gray level comentropy seal is Imgentropy。
Defining (x, y) is in entropy window sliding process, any one pixel coordinate that window center passes through, it is as the window
Central pixel point, if (i, j) ∈ space ΩθInterior pixel coordinate, f(i, j)Indicate the gray value of image changed the time.
The grayscale information entropy of corresponding center pixel:
ε=10E-6 is enabled, whereinP(i, j)Represent region ΩθThe gray space of upper each point is distributed
Situation:
P(i, j)=f(i, j)/N2
Wherein N2The size in the region is represented, i.e. (2M+1)2.It is this by area grayscale mean value as image information entropy in terms of
It calculates, the intensity profile situation of reflection that can be practical spatially.
(x, y)=(M+1, M+1) is defined when initial, local message entropy is calculated, with Z(x, y)Construct size and ImginIt is identical
Image local information entropy diagram fills entropy Z above (x, y) coordinate(x, y), obtain parts of images information entropy diagramAll coordinates (x, y), x ∈ (M+1, m- are successively traversed according to the movement at entropy window function center later
M), y ∈ (M+1, n-M), and remaining edge point, with 0 filling, obtain complete image information entropy diagram in information entropy diagram
Imgentropy。
Step 4.2: Img is schemed into fusionWTHEach pixel pixel value is as information entropy diagram ImgentropyWeight, will be local
Grayscale information entropy diagram ImgentropyImg is schemed with mergingWTHDoing the point multiplication operation of image, i.e. the pixel value of corresponding pixel points is multiplied, and
Weighting local gray level information entropy diagram Img is obtained after normalizationL_entropyAs infrared small target notable figure.
Step 5: calculate grayscale information entropy adaptive threshold:
The notable figure for acquiring infrared small target is calculated in step 4, this is also the most key part of entire algorithm.Depending on
Feel that the key link in the image partition method of conspicuousness is exactly to calculate the conspicuousness of each element in image, and calculate and provide significantly
Property adaptive threshold calculate, have reached the purpose of image segmentation.
Step 5.1: calculating infrared small target notable figure ImgL_entropyPixel value mean value Mean and pixel value variance Var,
And find out the maximum value H of wherein pixel valuemax；
Hmax=max { (gray(i, j)}
Wherein gray(i, j)Indicate infrared small target notable figure ImgL_entropyIn each pixel pixel value.
Step 5.2: in view of after obtaining infrared small target notable figure, contrast is lower, therefore carrying out binaryzation.It calculates
Threshold value T=α (Hmax-Mean)+βMean2+ ε Var, wherein α, β, ε are the normal number of setting,
Step 6: the threshold value T obtained using step 5 is to infrared small target notable figure ImgL_entropyBinary conversion treatment is carried out,
Pixel value is set to 0 lower than T in figure, is set to 1 not less than T；Obtain the binarization result of infrared small target notable figure
ImgL_entropy_BIt is distributed detection figure as final infrared small target, it is exactly infrared small target that wherein two-value, which turns to 1 region,.
Although the embodiments of the present invention has been shown and described above, it is to be understood that above-described embodiment is example
Property, it is not considered as limiting the invention, those skilled in the art are not departing from the principle of the present invention and objective
In the case where can make changes, modifications, alterations, and variations to the above described embodiments within the scope of the invention.
Claims (1)
1. a kind of weighted local entropy infrared small target detection method based on Multiscale Morphological Fusion, it is characterised in that: including
Following steps:
Step 1: one infrared image Img to be detected of input, image size are m*n；
Step 2: the image Img of input being converted into gray level image, obtains image Imgin；
Step 3: background inhibition is carried out using following steps:
Step 3.1: I morphology pre-processing structure unit m of settingi, 1≤i≤I, wherein I takes the integer greater than zero, and morphology
Pre-processing structure unit scale increases with i and is increased；
Step 3.2: selection morphology pre-processing structure unit miTo image ImginThe pretreatment of Top-Hat morphological images is carried out,
Pretreated image is obtained, WTHS is denoted asi；
Step 3.3: repeat step 3.2, until all settings morphology pre-processing structure unit all with image ImginIt completes
The pretreatment of Top-Hat morphological images, obtains multiple dimensioned Top-Hat pretreatment image collection WTHS；
Step 3.4: the gray value of adjacent two images corresponding pixel points in multiple dimensioned Top-Hat pretreatment image collection WTHS is made
Difference obtains disparity map:
WTHSTop_Hat(i+1)=WTHSi+1-WTHSi, 1≤i≤I-1
Step 3.5: repeating step 3.4, obtain the adjacent two images of I-1 group in multiple dimensioned Top-Hat pretreatment image collection WTHS
Between disparity map composition difference atlas WTHSTop_Hat；
Step 3.6: to difference atlas WTHSTop-HatThe fusion of element gray value minimum is carried out, blending image is obtained
WTHSTop-Hat-min；The gray average for calculating each image in multiple dimensioned Top-Hat pretreatment image collection WTHS, take it is therein most
Small mean value figure WTHSmean(min)；
Step 3.7: by image WTHSTop_Hat_minWith minimum mean figure WTHSmean(min)Middle corresponding pixel points gray value is added, and is obtained
Fusion figure Img after inhibiting to backgroundWTH；
Step 4: calculating image ImginImage local gray scale comentropy:
Step 4.1: choosing radius is the comentropy calculation window of M to image ImginFrom top to bottom, part ash is from left to right carried out
Spend comentropy operation: the center of comentropy calculation window is from image ImginThe position (M+1, M+1) start to slide, while calculate should
Local message entropy under window, to the last slides into image ImginThe position (m-M, n-M) until, image ImginMarginal information
Entropy is filled with 0, and obtaining local gray level comentropy seal is Imgentropy；
Step 4.2: by local gray level information entropy diagram ImgentropyImg is schemed with mergingWTHThe pixel value of corresponding pixel points is multiplied, and returns
Weighting local gray level information entropy diagram Img is obtained after one changeL-entropyAs infrared small target notable figure；
Step 5: calculate grayscale information entropy adaptive threshold:
Step 5.1: calculating infrared small target notable figure ImgL_entropyPixel value mean value Mean and pixel value variance Var, and look for
The wherein maximum value H of pixel value outmax；
Step 5.2: calculating threshold value T=α (Hmax-Mean)+βMean2+ ε Var, wherein α, β, ε are the normal number of setting；
Step 6: the threshold value T obtained using step 5 is to infrared small target notable figure ImgL_entropyBinary conversion treatment is carried out, in figure
Pixel value is set to 0 lower than T's, is set to 1 not less than T；Obtain the binarization result Img of infrared small target notable figureL_entropy_B
It is distributed detection figure as final infrared small target, it is exactly infrared small target that wherein two-value, which turns to 1 region,.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910017099.9A CN109816641B (en) | 2019-01-08 | 2019-01-08 | Multi-scale morphological fusion-based weighted local entropy infrared small target detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910017099.9A CN109816641B (en) | 2019-01-08 | 2019-01-08 | Multi-scale morphological fusion-based weighted local entropy infrared small target detection method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109816641A true CN109816641A (en) | 2019-05-28 |
| CN109816641B CN109816641B (en) | 2021-05-14 |
Family
ID=66604145
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910017099.9A Active CN109816641B (en) | 2019-01-08 | 2019-01-08 | Multi-scale morphological fusion-based weighted local entropy infrared small target detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109816641B (en) |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109903272A (en) * | 2019-01-30 | 2019-06-18 | 西安天伟电子系统工程有限公司 | Object detection method, device, equipment, computer equipment and storage medium |
| CN110223344A (en) * | 2019-06-03 | 2019-09-10 | 哈尔滨工程大学 | A kind of infrared small target detection method based on morphology and vision noticing mechanism |
| CN110807758A (en) * | 2019-09-10 | 2020-02-18 | 中国三峡建设管理有限公司 | Method, device, equipment and storage medium for detecting uncovered area of heat preservation quilt |
| CN110942437A (en) * | 2019-11-29 | 2020-03-31 | 石家庄铁道大学 | Adaptive top-hat transformation method based on Otsu-SSIM |
| CN110956180A (en) * | 2019-07-04 | 2020-04-03 | 中联重科股份有限公司 | Detection method and system of counterweight weight, acquisition method and system and crane |
| CN111275778A (en) * | 2020-01-08 | 2020-06-12 | 浙江省北大信息技术高等研究院 | Face sketch generating method and device |
| CN111783656A (en) * | 2020-06-30 | 2020-10-16 | 哈工程先进技术研究院(招远)有限公司 | Self-adaptive infrared small target detection method based on statistical characteristic pre-segmentation |
| CN112163606A (en) * | 2020-09-17 | 2021-01-01 | 哈尔滨工业大学（威海） | Infrared small target detection method based on block contrast weighting |
| CN112529896A (en) * | 2020-12-24 | 2021-03-19 | 山东师范大学 | Infrared small target detection method and system based on dark channel prior |
| CN113076888A (en) * | 2021-04-09 | 2021-07-06 | 中国电子科技集团公司第二十八研究所 | Multi-scale local binary model photoelectric image target detection method |
| CN113111878A (en) * | 2021-04-30 | 2021-07-13 | 中北大学 | Infrared weak and small target detection method under complex background |
| CN113160166A (en) * | 2021-04-16 | 2021-07-23 | 重庆飞唐网景科技有限公司 | Medical image data mining working method through convolutional neural network model |
| CN113421205A (en) * | 2021-07-16 | 2021-09-21 | 合肥工业大学 | Small target detection method combined with infrared polarization imaging |
| CN113827340A (en) * | 2021-09-09 | 2021-12-24 | 王其景 | Navigation system and method for managing information of surgical patient |
| CN114418913A (en) * | 2022-01-03 | 2022-04-29 | 中国电子科技集团公司第二十研究所 | ISAR and infrared image pixel level fusion method based on wavelet transformation |
| CN115205327A (en) * | 2022-05-18 | 2022-10-18 | 哈尔滨工业大学 | Infrared small target tracking method fusing historical library information |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103810709A (en) * | 2014-02-25 | 2014-05-21 | 南京理工大学 | Vessel-based registration method for eye fundus image and SD-OCT projection image |
| CN103839267A (en) * | 2014-02-27 | 2014-06-04 | 西安科技大学 | Building extracting method based on morphological building indexes |
| CN104156929A (en) * | 2014-09-05 | 2014-11-19 | 西安电子科技大学 | Infrared weak and small target background inhibiting method and device on basis of global filtering |
| CN104268844A (en) * | 2014-10-17 | 2015-01-07 | 中国科学院武汉物理与数学研究所 | Small target infrared image processing method based on weighing local image entropy |
| CN105469049A (en) * | 2015-11-24 | 2016-04-06 | 昆明理工大学 | Leakage sound emission signal identification method based on multi-scale morphological decomposition energy spectrum entropy and support vector machine |
| US20170224270A1 (en) * | 2016-02-10 | 2017-08-10 | Balter, Inc. | Optical transfer diagnosis for detection and monitoring of tissue disorders |
-
2019
- 2019-01-08 CN CN201910017099.9A patent/CN109816641B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103810709A (en) * | 2014-02-25 | 2014-05-21 | 南京理工大学 | Vessel-based registration method for eye fundus image and SD-OCT projection image |
| CN103839267A (en) * | 2014-02-27 | 2014-06-04 | 西安科技大学 | Building extracting method based on morphological building indexes |
| CN104156929A (en) * | 2014-09-05 | 2014-11-19 | 西安电子科技大学 | Infrared weak and small target background inhibiting method and device on basis of global filtering |
| CN104268844A (en) * | 2014-10-17 | 2015-01-07 | 中国科学院武汉物理与数学研究所 | Small target infrared image processing method based on weighing local image entropy |
| CN105469049A (en) * | 2015-11-24 | 2016-04-06 | 昆明理工大学 | Leakage sound emission signal identification method based on multi-scale morphological decomposition energy spectrum entropy and support vector machine |
| US20170224270A1 (en) * | 2016-02-10 | 2017-08-10 | Balter, Inc. | Optical transfer diagnosis for detection and monitoring of tissue disorders |
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109903272B (en) * | 2019-01-30 | 2021-09-03 | 西安天伟电子系统工程有限公司 | Target detection method, device, equipment, computer equipment and storage medium |
| CN109903272A (en) * | 2019-01-30 | 2019-06-18 | 西安天伟电子系统工程有限公司 | Object detection method, device, equipment, computer equipment and storage medium |
| CN110223344A (en) * | 2019-06-03 | 2019-09-10 | 哈尔滨工程大学 | A kind of infrared small target detection method based on morphology and vision noticing mechanism |
| CN110223344B (en) * | 2019-06-03 | 2023-09-29 | 哈尔滨工程大学 | Infrared small target detection method based on morphology and visual attention mechanism |
| CN110956180A (en) * | 2019-07-04 | 2020-04-03 | 中联重科股份有限公司 | Detection method and system of counterweight weight, acquisition method and system and crane |
| CN110807758A (en) * | 2019-09-10 | 2020-02-18 | 中国三峡建设管理有限公司 | Method, device, equipment and storage medium for detecting uncovered area of heat preservation quilt |
| CN110807758B (en) * | 2019-09-10 | 2022-11-08 | 中国三峡建设管理有限公司 | Method, device, equipment and storage medium for detecting uncovered area of heat preservation quilt |
| CN110942437A (en) * | 2019-11-29 | 2020-03-31 | 石家庄铁道大学 | Adaptive top-hat transformation method based on Otsu-SSIM |
| CN110942437B (en) * | 2019-11-29 | 2022-11-08 | 石家庄铁道大学 | Adaptive top-hat transformation method based on Otsu-SSIM |
| CN111275778A (en) * | 2020-01-08 | 2020-06-12 | 浙江省北大信息技术高等研究院 | Face sketch generating method and device |
| CN111275778B (en) * | 2020-01-08 | 2023-11-21 | 杭州未名信科科技有限公司 | Face simple drawing generation method and device |
| CN111783656A (en) * | 2020-06-30 | 2020-10-16 | 哈工程先进技术研究院(招远)有限公司 | Self-adaptive infrared small target detection method based on statistical characteristic pre-segmentation |
| CN111783656B (en) * | 2020-06-30 | 2024-03-08 | 哈尔滨工程大学 | Self-adaptive infrared small target detection method based on statistical characteristic pre-segmentation |
| CN112163606A (en) * | 2020-09-17 | 2021-01-01 | 哈尔滨工业大学（威海） | Infrared small target detection method based on block contrast weighting |
| CN112529896A (en) * | 2020-12-24 | 2021-03-19 | 山东师范大学 | Infrared small target detection method and system based on dark channel prior |
| CN113076888A (en) * | 2021-04-09 | 2021-07-06 | 中国电子科技集团公司第二十八研究所 | Multi-scale local binary model photoelectric image target detection method |
| CN113160166B (en) * | 2021-04-16 | 2022-02-15 | 宁波全网云医疗科技股份有限公司 | Medical image data mining working method through convolutional neural network model |
| CN113160166A (en) * | 2021-04-16 | 2021-07-23 | 重庆飞唐网景科技有限公司 | Medical image data mining working method through convolutional neural network model |
| CN113111878B (en) * | 2021-04-30 | 2022-03-18 | 中北大学 | Infrared weak and small target detection method under complex background |
| CN113111878A (en) * | 2021-04-30 | 2021-07-13 | 中北大学 | Infrared weak and small target detection method under complex background |
| CN113421205B (en) * | 2021-07-16 | 2022-11-15 | 合肥工业大学 | Small target detection method combined with infrared polarization imaging |
| CN113421205A (en) * | 2021-07-16 | 2021-09-21 | 合肥工业大学 | Small target detection method combined with infrared polarization imaging |
| CN113827340A (en) * | 2021-09-09 | 2021-12-24 | 王其景 | Navigation system and method for managing information of surgical patient |
| CN114418913A (en) * | 2022-01-03 | 2022-04-29 | 中国电子科技集团公司第二十研究所 | ISAR and infrared image pixel level fusion method based on wavelet transformation |
| CN114418913B (en) * | 2022-01-03 | 2024-04-19 | 中国电子科技集团公司第二十研究所 | ISAR and infrared image pixel level fusion method based on wavelet transformation |
| CN115205327A (en) * | 2022-05-18 | 2022-10-18 | 哈尔滨工业大学 | Infrared small target tracking method fusing historical library information |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109816641B (en) | 2021-05-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109816641A (en) | Weighted local entropy infrared small target detection method based on Multiscale Morphological Fusion | |
| Hou et al. | RISTDnet: Robust infrared small target detection network | |
| Xia et al. | Infrared small target detection based on multiscale local contrast measure using local energy factor | |
| Wu et al. | A double-neighborhood gradient method for infrared small target detection | |
| Wen et al. | A novel automatic change detection method for urban high-resolution remotely sensed imagery based on multiindex scene representation | |
| CN106952288B (en) | Based on convolution feature and global search detect it is long when block robust tracking method | |
| CN104408482B (en) | A kind of High Resolution SAR Images object detection method | |
| CN105957054B (en) | A kind of image change detection method | |
| CN110119728A (en) | Remote sensing images cloud detection method of optic based on Multiscale Fusion semantic segmentation network | |
| CN108986140A (en) | Target scale adaptive tracking method based on correlation filtering and color detection | |
| CN109272489A (en) | Inhibit the method for detecting infrared puniness target with multiple dimensioned local entropy based on background | |
| CN105427350B (en) | Color reproduction image altering detecting method based on local quaternary number Uniformly bounded | |
| CN105654516B (en) | Satellite image based on target conspicuousness is to ground weak moving target detection method | |
| Qi et al. | FTC-Net: Fusion of transformer and CNN features for infrared small target detection | |
| Cao et al. | Infrared small target detection based on derivative dissimilarity measure | |
| CN104268833A (en) | New image fusion method based on shift invariance shearlet transformation | |
| CN109215025A (en) | A kind of method for detecting infrared puniness target approaching minimization based on non-convex order | |
| Xue et al. | Unsupervised change detection using multiscale and multiresolution Gaussian-mixture-model guided by saliency enhancement | |
| CN101546429A (en) | Detection method of small target of feature space | |
| Zhang et al. | Infrared small target detection based on gradient correlation filtering and contrast measurement | |
| CN106056115B (en) | A kind of infrared small target detection method under non-homogeneous background | |
| Zhao et al. | Gesanet: Geospatial-awareness network for vhr remote sensing image change detection | |
| Xu et al. | The comparative study of three methods of remote sensing image change detection | |
| Jiang et al. | Automatic space debris extraction channel based on large field of view photoelectric detection system | |
| CN109087334A (en) | Target tracking method based on significant context rarefaction representation |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20230720 Address after: Room A601, National Digital Publishing Base, No. 996 Tiangu 7th Road, Gaoxin, Xi'an City, Shaanxi Province, 710000 Patentee after: XI'AN THUNDER DRIVEN ELECTRONIC TECHNOLOGY CO.,LTD. Address before: 710071 No. 2 Taibai South Road, Shaanxi, Xi'an Patentee before: XIDIAN University |
|
| TR01 | Transfer of patent right |