CN109492543A - The small target detecting method and system of infrared image - Google Patents
The small target detecting method and system of infrared image Download PDFInfo
- Publication number
- CN109492543A CN109492543A CN201811217578.7A CN201811217578A CN109492543A CN 109492543 A CN109492543 A CN 109492543A CN 201811217578 A CN201811217578 A CN 201811217578A CN 109492543 A CN109492543 A CN 109492543A
- Authority
- CN
- China
- Prior art keywords
- image
- background
- pixel
- infrared image
- small target
- 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
- 238000000034 method Methods 0.000 title claims abstract description 53
- 238000001914 filtration Methods 0.000 claims abstract description 82
- 238000012545 processing Methods 0.000 claims abstract description 30
- 230000005764 inhibitory process Effects 0.000 claims abstract description 18
- 230000008569 process Effects 0.000 claims description 12
- 238000000605 extraction Methods 0.000 claims description 3
- 230000002401 inhibitory effect Effects 0.000 claims description 3
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 230000036961 partial effect Effects 0.000 abstract description 4
- 230000006870 function Effects 0.000 description 20
- 238000001514 detection method Methods 0.000 description 12
- 230000008859 change Effects 0.000 description 8
- 230000000694 effects Effects 0.000 description 5
- 238000003860 storage Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000001629 suppression Effects 0.000 description 4
- 101100117236 Drosophila melanogaster speck gene Proteins 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 238000002474 experimental method Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 230000002708 enhancing effect Effects 0.000 description 2
- 238000003707 image sharpening Methods 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 238000003331 infrared imaging Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000005855 radiation Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 230000002829 reductive effect Effects 0.000 description 2
- 238000009416 shuttering Methods 0.000 description 2
- 229910003460 diamond Inorganic materials 0.000 description 1
- 239000010432 diamond Substances 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 230000008570 general process Effects 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 239000004615 ingredient Substances 0.000 description 1
- 230000000670 limiting effect Effects 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 210000003733 optic disk Anatomy 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000001902 propagating effect Effects 0.000 description 1
- 238000011897 real-time detection Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- 230000003313 weakening effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G06T5/73—
-
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/07—Target detection
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
The present invention relates to field of image processings, are related to a kind of small target detecting method of infrared image, comprising: are handled using median filtering technology original infrared image；Substracting background inhibition processing is carried out to the image after original infrared image and median filtering using background differential technique；Treated image is inhibited to carry out Grads Sharp processing using preset threshold value substracting background；The infrared small target in image after extracting Grads Sharp.The present invention also proposes the small target deteection system of infrared image.The present invention carries out median filtering to original infrared image to filter out partial noise clutter, that realizes image smoothly obtains background component, then the image after median filtering is done into substracting background inhibition processing with original infrared image, it is finally sharpened using gradient method and threshold restriction is carried out to residual image, to finally detect target.The present invention can effectively filter out the background clutter in image, highlight target, so that signal-to-noise ratio is remarkably reinforced, and operation is simply easily achieved.
Description
Technical field
The present invention relates to field of image processing, in particular to the small target detecting method and system of a kind of infrared image.
Background technique
Infrared imaging system is compared with VISIBLE LIGHT SYSTEM, and environmental suitability is better than visible light, especially at night and evil
It is even more so under bad weather；Compared with radar system, infrared wavelength is shorter, can obtain having very high-resolution target figure
Picture.Therefore, the energy of target itself radiation is captured and tracked using infrared detector to realize accurate guidance technology in the modern times
Increasingly important role is played in science and techniques of defence.And successfully infrared imaging guidance system must be equipped with it is efficient, accurate
Infrared small target detection device, this is of great significance for finding and tracking suspicious target in time.
Infrared small target detection refers to that the relative position due to imaging system distance objective, imaging area is only several farther out
A pixel shows as acnode or spot to more than ten of pixel in visual field, can not show its own shape, amplitude distribution etc.
The detection of the Small object of feature.The characteristics of due to Small object itself, along with the complexity of background, its detection and tracking is opposite
The more of difficulty are wanted in Area Objects and big target.
Currently based on the small target detecting method of infrared image mainly there are two important developing direction: Detect before Track
(Detect Before Track, abbreviation DBT) and root-first search (Track Before Detect, abbreviation TBD).
DBT refers to that the intensity detection that object pixel is first passed through before tracking target goes out target, and then is repaired by certain decision criteria
Positive testing result finally obtains position and the motion profile of target, realizes real-time detection and tracking to target object.It is general
Process is: eliminating background signal and noise clutter in original image using background suppression technology first, obtains potential target point；
Potential target point is judged then according to one decision rule of property settings of target in image, and according to this criterion, is picked
Except false-alarm point；Finally the centre coordinate of the target finally confirmed is exported, to realize the effect of tracking.Compared with TBD, DBT
It is easy to operate, it is easy to accomplish, it is suitable for the higher image detection of signal-to-noise ratio.
But these types of detection method can not be rejected when aerial there are cloud noise or there are when surface feature background interference
Background clutter simultaneously highlights target, and then disturbs the detection of real goal.
Summary of the invention
Embodiments of the present invention aim to solve at least one of the technical problems existing in the prior art.For this purpose, of the invention
Embodiment need to provide the small target detecting method and system of a kind of infrared image.
A kind of small target detecting method of infrared image of embodiment of the present invention characterized by comprising
Step 1, original infrared image is handled using median filtering technology；
Step 2, substracting background suppression is carried out to the image after original infrared image and median filtering using background differential technique
System processing；
Step 3, treated image is inhibited to carry out Grads Sharp processing using preset threshold value substracting background；
Step 4, the infrared small target in image after extracting Grads Sharp.
In a kind of embodiment, step 1 includes:
Step 11, the center for the default template chosen is overlapped with the pixel being selected in original infrared image, and
The template is successively slipped in the images；
Step 12, the gray value for the corresponding pixel points that template slips in image is read；
Step 13, the gray value of reading is ranked up；
Step 14, most intermediate gray value is chosen from ranking results according to median filtering principle；
Step 15, the gray value of taking-up is assigned to the pixel of corresponding templates center in image, traversal whole image is complete
At median filtering process.
In a kind of embodiment, step 2 includes:
Step 21, the coordinate of all corresponding pixel points and each in the image after obtaining original infrared image and median filtering
The current frame image of a pixel；
Step 22, smothing filtering is carried out to current frame image, forms background frames image；
Step 23, substracting background inhibition processing is carried out to current frame image and background frames image according to the following formula, obtained
Residual image:
dk(x, y)=| fk(x,y)-bk(x,y)|
Wherein, (x, y) is pixel coordinate, fk(x, y) is current frame image, bk(x, y) is the background image of present frame,
dk(x, y) is the difference result of present frame and background frames；
Step 24, secondary filtering is carried out to residual image and exports the image after secondary filtering.
In a kind of embodiment, step 3 includes:
Step 31, the calculating of gradient value is carried out according to following formula:
Wherein, f (x, y) indicates image function,Indicate the gradient magnitude at pixel (x, y)；
Step 32, according to the gradient value of calculating choose preset threshold value the substracting background image that inhibits that treated is carried out it is sharp
Change.
In a kind of embodiment, step 3 includes:
Step 31, the calculating of gradient value is carried out according to following formula:
Wherein, f (x, y) indicates image function,Indicate the gradient magnitude at pixel (x, y)；
Step 32, according to the gradient value of calculating choose preset threshold value the substracting background image that inhibits that treated is carried out it is sharp
Change.
Present embodiment also proposes a kind of small target deteection system of infrared image characterized by comprising
Medium filtering module, for being handled using median filtering technology original infrared image；
Background difference block, for being carried out using background differential technique to the image after original infrared image and median filtering
Substracting background inhibition processing；
Grads Sharp module, for inhibiting treated image to carry out Grads Sharp using preset threshold value substracting background
Processing；
Small object extraction module, for extracting the infrared small target in the image after Grads Sharp.
In a kind of embodiment, medium filtering module includes:
Coincidence unit, the center of the default template for that will choose and the pixel weight being selected in original infrared image
It closes, and successively slips over the template in the images；
Reading unit, for reading the gray value for the corresponding pixel points that template in image slips over；
Sequencing unit, for being ranked up to the gray value of reading；
Median cells, for choosing most intermediate gray value from ranking results according to median filtering principle；
Traversal Unit, for the gray value of taking-up to be assigned to the pixel of corresponding templates center in image, traversal is entire
Image completes median filtering process.
In a kind of embodiment, background difference block includes:
Current frame unit, for obtaining the seat of all corresponding pixel points in the image after original infrared image and median filtering
The current frame image of mark and each pixel；
Background frame unit forms background frames image for carrying out smothing filtering to current frame image；
Inhibit unit, for carrying out at substracting background inhibition to current frame image and background frames image according to the following formula
Reason obtains residual image:
dk(x, y)=| fk(x,y)-bk(x,y)|
Wherein, (x, y) is pixel coordinate, fk(x, y) is current frame image, bk(x, y) is the background image of present frame,
dk(x, y) is the difference result of present frame and background frames；
Secondary filtering unit, for carrying out secondary filtering to residual image and exporting the image after secondary filtering.
In a kind of embodiment, Grads Sharp module includes:
Computing unit, for carrying out the calculating of gradient value according to following formula:
Wherein, f (x, y) indicates image function,Indicate the gradient magnitude at pixel (x, y)；
Unit is sharpened, for choosing preset threshold value to the substracting background image that inhibits that treated according to the gradient value of calculating
It is sharpened.
In a kind of embodiment, Grads Sharp module includes:
Computing unit, for carrying out the calculating of gradient value according to following formula:
Wherein, f (x, y) indicates image function,Indicate the gradient magnitude at pixel (x, y)；
Unit is sharpened, for choosing preset threshold value to the substracting background image that inhibits that treated according to the gradient value of calculating
It is sharpened.
The small target detecting method and system of the infrared image of embodiment of the present invention first carry out original infrared image
Median filtering is to filter out partial noise clutter, and that realizes image smoothly obtains background component, then by the figure after median filtering
As doing substracting background inhibition processing with original infrared image, finally sharpened using gradient method to residual image progress threshold restriction,
To finally detect target.The present invention can effectively filter out the background clutter in infrared image, highlight target, so that noise
Than being remarkably reinforced, and operation is simple, it is easy to accomplish.
The advantages of additional aspect 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
The above-mentioned and/or additional aspect and advantage of embodiments of the present invention are from combination following accompanying drawings to embodiment
It will be apparent and be readily appreciated that in description, in which:
Fig. 1 is the flow diagram of the small target detecting method of the infrared image of embodiment of the present invention；
Fig. 2 is the template schematic diagram of the median filtering of embodiment of the present invention；
Fig. 3 is that different threshold values are shown in image gradient Edge contrast when target is outside cloud layer in embodiment of the present invention
It is intended to；
Fig. 4 is different threshold values in image gradient Edge contrast when target part is by cloud cover in embodiment of the present invention
Schematic diagram；
Fig. 5 is the composition schematic diagram of the small target deteection system of the infrared image of embodiment of the present invention.
Specific embodiment
Embodiments of the present invention are described below in detail, the example of embodiment is shown in the accompanying drawings, wherein identical or class
As label indicate same or similar element or element with the same or similar functions from beginning to end.Below with reference to attached
The embodiment of figure description is exemplary, and can only be used to explain embodiments of the present invention, and should not be understood as to the present invention
Embodiment limitation.
Embodiment 1
Referring to Fig. 1, the small target detecting method of the infrared image of embodiment of the present invention, comprising:
Step 1, original infrared image is handled using median filtering technology.
Step 2, substracting background suppression is carried out to the image after original infrared image and median filtering using background differential technique
System processing.
Step 3, treated image is inhibited to carry out Grads Sharp processing using preset threshold value substracting background.
Step 4, the infrared small target in image after extracting Grads Sharp.
In the present embodiment, infrared image is obtained using ThermoproTMTP8 thermal infrared imager, the infrared image of acquisition is big
Small is 288 × 348 pixels.
In step 1, comprising:
Step 11, the center for the default template chosen is overlapped with the pixel being selected in original infrared image, and
The template is successively slipped in the images.
Step 12, the gray value for the corresponding pixel points that template slips in image is read.
Step 13, the gray value of reading is ranked up.
Step 14, most intermediate gray value is chosen from ranking results according to median filtering principle.
Step 15, the gray value of taking-up is assigned to the pixel of corresponding templates center in image, traversal whole image is complete
At median filtering process.
Median filtering technology is a kind of nonlinear smoothing filter technology based on sort method theory, it is by each pixel
Gray value be set as the intermediate value of the gray value of all pixels point in the point field.Because median filter is unwise to exceptional value
Sense, so median filtering can reduce influence of the exceptional value to it in the case where not weakening picture contrast, and is able to maintain red
The integrality of outer image section marginal information saves the part detailed information of signal.
Two dimension median filter be central point grey scale pixel value is replaced with the gray scale intermediate value of pixel N number of in field, if
The size for filtering neighborhood is A=N × N (N is usually odd number), then has:
xij=MedianXij(m,n)≠i,j
Wherein, (i, j) is pixel coordinate, Xi,j(m, n) indicates the pixel in the median filtering window of pixel (i, j)
Point, Median expression take median operation.
The process of median filtering is: the template of one filtering of setting adopts the picture signal to be treated of input
Sample, arbitrarily chooses one of pixel as the center of the template of setting, then around the center be arranged again one it is determining
Field sorts each pixel according to gray value size in this field, chooses most intermediate gray value as the pixel
Output gray level value, traversal whole image filtering can be completed.
The common template of two dimension median filter has: cross, and diamond shape is rectangular and round etc..
As shown in Fig. 2, the present invention finally uses circular shuttering by many experiments.
Since median filtering technology is a kind of typical nonlinear filtering technique, so in practical applications, filter
The operation of template and image to be processed is no longer mask convolution, but template sorts.Template sequence refer to template extract to
Handle image subset identical with template size in image, and the calculating process that pixel therein is sorted according to its amplitude.
In step 2, comprising:
Step 21, the coordinate of all corresponding pixel points and each in the image after obtaining original infrared image and median filtering
The current frame image of a pixel.
Step 22, smothing filtering is carried out to current frame image, forms background frames image.
Step 23, substracting background inhibition processing is carried out to current frame image and background frames image according to the following formula, obtained
Residual image:
dk(x, y)=| fk(x,y)-bk(x,y)|
Wherein, (x, y) is pixel coordinate, fk(x, y) is current frame image, bk(x, y) is the background image of present frame,
dk(x, y) is the difference result of present frame and background frames.
Step 24, to residual image dk(x, y) carries out median filtering again and exports filtered image.
Background in infrared image is mainly the slowly varying low frequency part of large area, and details ingredient is less, in background
Pixel gray value and the gray value correlation of its surrounding pixel point are very strong, and the gray value on Weak target point and picture around it
The gray value correlation of vegetarian refreshments is very weak, and target point has thermal radiation property more stronger than background, and gray value is often in part
Maximum is shown as in region.
In view of above-mentioned characteristic, the coordinate of all corresponding pixel points in the image after obtaining original infrared image and median filtering
After the current frame image of each pixel, smothing filtering is carried out to current frame image, background frames image is formed, further according to upper
Current frame image and background frames image are carried out substracting background inhibition processing by the formula of step 23 in text, so that background pixel point phase
It is smaller to subtract rear residual error, and to subtract each other rear residual error larger for target pixel points, to effectively inhibit background signal, further increases image
Signal to noise ratio.Meanwhile the echo signal in image also receives different degrees of limitation, therefore after needing to handle substracting background inhibition
Obtained residual image carries out secondary median filtering, to enhance echo signal, to detect Small object.
In step 3, the purpose of image sharpening is enhancing image border, keeps the edge of target object distinct, in order to extract
Target object.Substantially, the fuzzy of image border is average calculating operation or integral operation as a result, and its inverse operation such as differential is transported
Calculation (solving the change rate of signal) can then reinforce high fdrequency component and fuzzy echo signal is made to become clear.
The present embodiment is sharpened processing to image using gradient method.
For image function f (x, y), its gradient at (x, y) is a vector, positioning are as follows:
The amplitude of gradientIt can be calculated by following formula:
From the above equation, we can see that the numerical value of gradient be exactly in function f (x, y) in the unit distance on its maximum rate of change direction
The increased amount of institute.The present embodiment includes two kinds of situations for the calculating of gradient value:
In a kind of situation, for discrete picture processing, the size of gradient is commonly used, and micro- come approximate single order with difference
PointWithTo which step 3 includes:
Step 31, the calculating of gradient value is carried out according to following formula:
Wherein, f (x, y) indicates image function,Indicate the gradient magnitude at pixel (x, y)；
Step 32, according to the gradient value of calculating choose preset threshold value the substracting background image that inhibits that treated is carried out it is sharp
Change.
In another situation, in order to improve calculating speed, simplified calculation method is calculated using horizontal vertical calculus of finite differences,
To which step 3 includes:
Step 31, the calculating of gradient value is carried out according to following formula:
Wherein, f (x, y) indicates image function,Indicate the gradient magnitude at pixel (x, y)；
Step 32, according to the gradient value of calculating choose preset threshold value the substracting background image that inhibits that treated is carried out it is sharp
Change.
In the present embodiment, Grads Sharp part is to be shown by two gray levels of setting to complete image, that is, uses two-value
Image is shown that this method can very easily analyze the edge position information of image, and display methods is as follows:
Optimal threshold T should both make target highlight, and also answer the number of the reduction false-alarm point of maximum possible.If value is too
Small, image will appear many false-alarm points；If value is too big, target can be flooded by background, can not be identified.
Emulation experiment is carried out for two kinds of infrared small target image sequences under different background, wherein sequence 1 is target
Not by the image of cloud cover, sequence 2 is the image containing target part by cloud cover.
As shown in figure 3, infrared small target is in outside cloud layer in the 27th frame image of sequence 1, when threshold value T is less than 24
When not only detected target, also detected that point brighter in background image, i.e. false-alarm point.When such as threshold value T=20, it can see
Into original image, the speck in left side is detected, and affects detection effect.Threshold value can be increased at this time, i.e., when T is in
When between 24 and 60, can effectively detect target and remove false-alarm point, if but threshold value choose it is excessive, target information also can
It is eliminated, so that target can not be detected.
As shown in figure 4, choosing the 178th frame image of sequence 2, it can be seen that infrared small target will enter cloud layer at this time,
It is covered by part cloud layer information.Compared with the detection effect of the 178th frame, the range of threshold value T is obviously reduced at this time, i.e. T's
Infrared small target can be detected when value is between 21 and 31, the redundancy speck in image will be detected when less than 21.
The present embodiment carries out median filtering to original infrared image first to filter out partial noise clutter, realizes image
Background component smoothly is obtained, the image after median filtering is then done into substracting background inhibition processing with original infrared image, finally
It is sharpened using gradient method and threshold restriction is carried out to residual image, to finally detect target.The present invention can effectively filter out red
Background clutter in outer image, highlights target, so that signal-to-noise ratio is remarkably reinforced, and operation is simple, it is easy to accomplish.
Embodiment 2
As shown in figure 5, the present embodiment also proposes a kind of small target deteection system of infrared image, comprising:
Medium filtering module, for being handled using median filtering technology original infrared image.
Background difference block, for being carried out using background differential technique to the image after original infrared image and median filtering
Substracting background inhibition processing.
Grads Sharp module, for inhibiting treated image to carry out Grads Sharp using preset threshold value substracting background
Processing.
Small object extraction module, for extracting the infrared small target in the image after Grads Sharp.
In the present embodiment, infrared image is obtained using ThermoproTMTP8 thermal infrared imager, the infrared image of acquisition is big
Small is 288 × 348 pixels.
Specifically, medium filtering module includes:
Coincidence unit, the center of the default template for that will choose and the pixel weight being selected in original infrared image
It closes, and successively slips over the template in the images.
Reading unit, for reading the gray value for the corresponding pixel points that template in image slips over.
Sequencing unit, for being ranked up to the gray value of reading.
Median cells, for choosing most intermediate gray value from ranking results according to median filtering principle.
Traversal Unit, for the gray value of taking-up to be assigned to the pixel of corresponding templates center in image, traversal is entire
Image completes median filtering process.
The present embodiment also uses the circular shuttering such as Fig. 2.
The work engagement process of each unit can be carried out in medium filtering module with reference implementation example 1, no longer superfluous herein
It states.
Specifically, background difference block includes:
Current frame unit, for obtaining the seat of all corresponding pixel points in the image after original infrared image and median filtering
The current frame image of mark and each pixel.
Background frame unit forms background frames image for carrying out smothing filtering to current frame image.
Inhibit unit, for carrying out at substracting background inhibition to current frame image and background frames image according to the following formula
Reason obtains residual image:
dk(x, y)=| fk(x,y)-bk(x,y)|
Wherein, (x, y) is pixel coordinate, fk(x, y) is current frame image, bk(x, y) is the background image of present frame,
dk(x, y) is the difference result of present frame and background frames.
Secondary filtering unit, for carrying out secondary filtering to residual image and exporting the image after secondary filtering.
I.e. background difference block using pixel gray value and its surrounding pixel point in background gray value correlation very
This very weak characteristic of gray value correlation of gray value on Weak target point and its surrounding pixel point by force, first passes through present frame
Unit obtains current frame image, then forms background frames image by background frame unit, then carries out substracting background suppression by inhibition unit
System processing obtains residual image, finally carries out secondary filtering to residual image by secondary filtering unit, to enhance echo signal, from
And detect Small object.
The work engagement process of each unit can be carried out in background difference block with reference implementation example 1, no longer superfluous herein
It states.
The purpose that Grads Sharp module carries out image sharpening is enhancing image border, keeps the edge of target object distinct, with
Convenient for extracting target object.Substantially, the fuzzy of image border is average calculating operation or integral operation as a result, and its inverse operation
It can reinforce high fdrequency component as differentiating if (solve signal change rate) fuzzy echo signal is made to become clear.
The present embodiment is sharpened processing to image using gradient method.
For image function f (x, y), its gradient at (x, y) is a vector, positioning are as follows:
The amplitude of gradientIt can be calculated by following formula:
From the above equation, we can see that the numerical value of gradient be exactly in function f (x, y) in the unit distance on its maximum rate of change direction
The increased amount of institute.The present embodiment includes two kinds of situations for the calculating of gradient value:
In a kind of situation, for discrete picture processing, the size of gradient is commonly used, and micro- come approximate single order with difference
PointWithTo which Grads Sharp module includes:
Computing unit, for carrying out the calculating of gradient value according to following formula:
Wherein, f (x, y) indicates image function,Indicate the gradient magnitude at pixel (x, y).
Unit is sharpened, for choosing preset threshold value to the substracting background image that inhibits that treated according to the gradient value of calculating
It is sharpened.
In another situation, in order to improve calculating speed, simplified calculation method is calculated using horizontal vertical calculus of finite differences,
To which Grads Sharp module includes:
Computing unit, for carrying out the calculating of gradient value according to following formula:
Wherein, f (x, y) indicates image function,Indicate the gradient magnitude at pixel (x, y)；
Unit is sharpened, for choosing preset threshold value to the substracting background image that inhibits that treated according to the gradient value of calculating
It is sharpened.
In the present embodiment, Grads Sharp part is to be shown by two gray levels of setting to complete image, that is, uses two-value
Image is shown that this method can very easily analyze the edge position information of image, and display methods is as follows:
Optimal threshold T should both make target highlight, and also answer the number of the reduction false-alarm point of maximum possible.If value is too
Small, image will appear many false-alarm points；If value is too big, target can be flooded by background, can not be identified.
Emulation experiment is carried out for two kinds of infrared small target image sequences under different background, wherein sequence 1 is that have mesh
For mark not by the image of cloud cover, sequence 2 is target part by the image of cloud cover.
As shown in figure 3, infrared small target is in outside cloud layer in the 27th frame image of sequence 1, when threshold value T is less than 24
When not only detected target, also detected that point brighter in background image, i.e. false-alarm point.When such as threshold value T=20, it can see
Into original image, the speck in left side is detected, and affects detection effect.Threshold value can be increased at this time, i.e., when T is in
When between 24 and 60, can effectively detect target and remove false-alarm point, if but threshold value choose it is excessive, target information also can
It is eliminated, so that target can not be detected.
As shown in figure 4, choosing the 178th frame image of sequence 2, it can be seen that infrared small target will enter cloud layer at this time,
Part has been covered by cloud layer information.Compared with the detection effect of the 27th frame, the range of threshold value T is obviously reduced at this time, i.e. T's takes
Infrared small target can be detected when value is between 21 and 31, will detect that the redundancy in image is bright when less than 21 or greater than 31
Spot.
The present embodiment carries out median filtering to original infrared image first to filter out partial noise clutter, realizes image
Background component smoothly is obtained, the image after median filtering is then done into substracting background inhibition processing with original infrared image, finally
It is sharpened using gradient method and threshold restriction is carried out to residual image, to finally detect target.The present invention can effectively filter out red
Background clutter in outer image, highlights target, so that signal-to-noise ratio is remarkably reinforced, and operation is simple, it is easy to accomplish.
In the description of this specification, reference term " embodiment ", " some embodiments ", " schematically implementation
The description of mode ", " example ", specific examples or " some examples " etc. means the tool described in conjunction with the embodiment or example
Body characteristics, structure, material or feature are contained at least one embodiment or example of the invention.In the present specification,
Schematic expression of the above terms are not necessarily referring to identical embodiment or example.Moreover, the specific features of description, knot
Structure, material or feature can be combined in any suitable manner in any one or more embodiments or example.
Any process described otherwise above or method description are construed as in flow chart or herein, and expression includes
It is one or more for realizing specific logical function or process the step of executable instruction code module, segment or portion
Point, and the range of the preferred embodiment of the present invention includes other realization, wherein can not press shown or discussed suitable
Sequence, including according to related function by it is basic simultaneously in the way of or in the opposite order, Lai Zhihang function, this should be of the invention
Embodiment person of ordinary skill in the field understood.
Expression or logic and/or step described otherwise above herein in flow charts, for example, being considered use
In the order list for the executable instruction for realizing logic function, may be embodied in any computer-readable medium, for
Instruction execution system, device or equipment (such as computer based system, including the system of processing module or other can be from instruction
Execute system, device or equipment instruction fetch and the system that executes instruction) use, or combine these instruction execution systems, device or
Equipment and use.For the purpose of this specification, " computer-readable medium " can be it is any may include, store, communicating, propagating or
Transfer program uses for instruction execution system, device or equipment or in conjunction with these instruction execution systems, device or equipment
Device.The more specific example (non-exhaustive list) of computer-readable medium include the following: there are one or more wirings
Electrical connection section (electronic device), portable computer diskette box (magnetic device), random access memory (RAM), read-only memory
(ROM), erasable edit read-only storage (EPROM or flash memory), fiber device and portable optic disk is read-only deposits
Reservoir (CDROM).In addition, computer-readable medium can even is that the paper that can print described program on it or other are suitable
Medium, because can then be edited, be interpreted or when necessary with it for example by carrying out optical scanner to paper or other media
His suitable method is handled electronically to obtain described program, is then stored in computer storage.
It should be appreciated that each section of embodiments of the present invention can be with hardware, software, firmware or their combination come real
It is existing.In the above-described embodiment, multiple steps or method can be with storages in memory and by suitable instruction execution system
The software or firmware of execution is realized.For example, if realized with hardware, in another embodiment, ability can be used
Any one of following technology or their combination well known to domain is realized: being had for realizing logic function to data-signal
The discrete logic of logic gates, the specific integrated circuit with suitable combinational logic gate circuit, programmable gate array
(PGA), field programmable gate array (FPGA) etc..
Those skilled in the art are understood that realize all or part of step that above-described embodiment method carries
It suddenly is that relevant hardware can be instructed to complete by program, the program can store in a kind of computer-readable storage medium
In matter, which when being executed, includes the steps that one or a combination set of embodiment of the method.
In addition, each functional unit in various embodiments of the present invention can integrate in a processing module, it can also
To be that each unit physically exists alone, can also be integrated in two or more units in a module.It is above-mentioned integrated
Module both can take the form of hardware realization, can also be realized in the form of software function module.The integrated module
If in the form of software function module realize and when sold or used as an independent product, also can store one calculating
In machine read/write memory medium.
Storage medium mentioned above can be read-only memory, disk or CD etc..
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 within the scope of the invention can be to above-mentioned
Embodiment is changed, modifies, replacement and variant.
Claims (10)
1. a kind of small target detecting method of infrared image characterized by comprising
Step 1, original infrared image is handled using median filtering technology；
Step 2, the image after original infrared image and median filtering is carried out at substracting background inhibition using background differential technique
Reason；
Step 3, treated image is inhibited to carry out Grads Sharp processing using preset threshold value substracting background；
Step 4, the infrared small target in image after extracting Grads Sharp.
2. the small target detecting method of infrared image as described in claim 1, which is characterized in that step 1 includes:
Step 11, the center for the default template chosen is overlapped with the pixel being selected in original infrared image, and at this
The template is successively slipped in image；
Step 12, the gray value for the corresponding pixel points that template slips in image is read；
Step 13, the gray value of reading is ranked up；
Step 14, most intermediate gray value is chosen from ranking results according to median filtering principle；
Step 15, the gray value of taking-up is assigned to the pixel of corresponding templates center in image, is traversed in whole image completion
Value filtering process.
3. the small target detecting method of infrared image as described in claim 1, which is characterized in that step 2 includes:
Step 21, the coordinate of all corresponding pixel points and each picture in the image after obtaining original infrared image and median filtering
The current frame image of vegetarian refreshments；
Step 22, smothing filtering is carried out to current frame image, forms background frames image；
Step 23, substracting background inhibition processing is carried out to current frame image and background frames image according to the following formula, obtains residual error
Image:
dk(x, y)=| fk(x,y)-bk(x,y)|
Wherein, (x, y) is pixel coordinate, fk(x, y) is current frame image, bk(x, y) is the background image of present frame, dk(x,
Y) be present frame and background frames difference result；
Step 24, secondary filtering is carried out to residual image and exports the image after secondary filtering.
4. the small target detecting method of infrared image as described in claim 1, which is characterized in that step 3 includes:
Step 31, the calculating of gradient value is carried out according to following formula:
| ▽ f (x, y) |={ [f (x, y)-f (x+1, y)]2+[f(x,y)-f(x,y+1)]2}12
Wherein, f (x, y) indicates image function, | ▽ f (x, y) | indicate the gradient magnitude at pixel (x, y)；
Step 32, preset threshold value is chosen according to the gradient value of calculating to inhibit that treated to substracting background image is sharpened.
5. the small target detecting method of infrared image as described in claim 1, which is characterized in that step 3 includes:
Step 31, the calculating of gradient value is carried out according to following formula:
| ▽ f (x, y) |=| f (x, y)-f (x+1, y) |+| f (x, y)-f (x, y+1) |
Wherein, f (x, y) indicates image function, | ▽ f (x, y) | indicate the gradient magnitude at pixel (x, y)；
Step 32, preset threshold value is chosen according to the gradient value of calculating to inhibit that treated to substracting background image is sharpened.
6. a kind of small target deteection system of infrared image characterized by comprising
Medium filtering module, for being handled using median filtering technology original infrared image；
Background difference block, for carrying out difference to the image after original infrared image and median filtering using background differential technique
Background inhibition processing；
Grads Sharp module, for inhibiting treated image to carry out at Grads Sharp using preset threshold value substracting background
Reason；
Small object extraction module, for extracting the infrared small target in the image after Grads Sharp.
7. the small target deteection system of infrared image as claimed in claim 6, which is characterized in that medium filtering module includes:
Coincidence unit, for the center for the default template chosen to be overlapped with the pixel being selected in original infrared image,
And the template is successively slipped in the images；
Reading unit, for reading the gray value for the corresponding pixel points that template in image slips over；
Sequencing unit, for being ranked up to the gray value of reading；
Median cells, for choosing most intermediate gray value from ranking results according to median filtering principle；
Traversal Unit traverses whole image for the gray value of taking-up to be assigned to the pixel of corresponding templates center in image
Complete median filtering process.
8. the small target deteection system of infrared image as claimed in claim 6, which is characterized in that background difference block includes:
Current frame unit, for obtain in the image after original infrared image and median filtering the coordinate of all corresponding pixel points and
The current frame image of each pixel；
Background frame unit forms background frames image for carrying out smothing filtering to current frame image；
Inhibit unit, for carrying out substracting background inhibition processing to current frame image and background frames image according to the following formula, obtains
Obtain residual image:
dk(x, y)=| fk(x,y)-bk(x,y)|
Wherein, (x, y) is pixel coordinate, fk(x, y) is current frame image, bk(x, y) is the background image of present frame, dk(x,
Y) be present frame and background frames difference result；
Secondary filtering unit, for carrying out secondary filtering to residual image and exporting the image after secondary filtering.
9. the small target deteection system of infrared image as claimed in claim 6, which is characterized in that Grads Sharp module includes:
Computing unit, for carrying out the calculating of gradient value according to following formula:
| ▽ f (x, y) |={ [f (x, y)-f (x+1, y)]2+[f(x,y)-f(x,y+1)]2}1/2
Wherein, f (x, y) indicates image function, | ▽ f (x, y) | indicate the gradient magnitude at pixel (x, y)；
Unit is sharpened, for choosing preset threshold value according to the gradient value of calculating treated is inhibited to substracting background image carries out
It sharpens.
10. the small target deteection system of infrared image as claimed in claim 6, which is characterized in that Grads Sharp module includes:
Computing unit, for carrying out the calculating of gradient value according to following formula:
| ▽ f (x, y) |=| f (x, y)-f (x+1, y) |+| f (x, y)-f (x, y+1) |
Wherein, f (x, y) indicates image function, | ▽ f (x, y) | indicate the gradient magnitude at pixel (x, y)；
Unit is sharpened, for choosing preset threshold value according to the gradient value of calculating treated is inhibited to substracting background image carries out
It sharpens.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811217578.7A CN109492543A (en) | 2018-10-18 | 2018-10-18 | The small target detecting method and system of infrared image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811217578.7A CN109492543A (en) | 2018-10-18 | 2018-10-18 | The small target detecting method and system of infrared image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109492543A true CN109492543A (en) | 2019-03-19 |
Family
ID=65691959
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811217578.7A Pending CN109492543A (en) | 2018-10-18 | 2018-10-18 | The small target detecting method and system of infrared image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109492543A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110111261A (en) * | 2019-03-28 | 2019-08-09 | 福州瑞芯微电子股份有限公司 | Adaptive equalization processing method, electronic equipment and the computer readable storage medium of image |
| CN110135344A (en) * | 2019-05-16 | 2019-08-16 | 电子科技大学 | Based on the method for detecting infrared puniness target for weighting fixed order expression |
| CN110276788A (en) * | 2019-06-12 | 2019-09-24 | 北京轩宇空间科技有限公司 | Method and apparatus for infrared imaging formula target seeker target following |
| CN110400294A (en) * | 2019-07-18 | 2019-11-01 | 湖南宏动光电有限公司 | A kind of Infrared Target Detection system and detection method |
| CN111079546A (en) * | 2019-11-22 | 2020-04-28 | 重庆师范大学 | Unmanned aerial vehicle pest detection method |
| CN111337133A (en) * | 2020-03-02 | 2020-06-26 | 浙江大华技术股份有限公司 | Infrared data generation method and device and infrared data analysis method and device |
| CN111401195A (en) * | 2020-03-10 | 2020-07-10 | 上海航天控制技术研究所 | Sea surface target detection method based on multiband infrared image |
| CN111428573A (en) * | 2020-03-02 | 2020-07-17 | 南京莱斯电子设备有限公司 | Infrared weak and small target detection false alarm suppression method under complex background |
| CN112954138A (en) * | 2021-02-20 | 2021-06-11 | 东营市阔海水产科技有限公司 | Aquatic economic animal image acquisition method, terminal equipment and movable material platform |
| CN113613007A (en) * | 2021-07-19 | 2021-11-05 | 青岛信芯微电子科技股份有限公司 | Three-dimensional color lookup table generation method and display device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1514408A (en) * | 2002-09-12 | 2004-07-21 | 上海交通大学 | Infra red detecting and tracing method for weak target under complex background condition |
| CN103310419A (en) * | 2013-06-18 | 2013-09-18 | 深圳大学 | Small target image processing device |
| US8724850B1 (en) * | 2011-06-21 | 2014-05-13 | The United States Of America As Represented By The Secretary Of The Navy | Small object detection using meaningful features and generalized histograms |
| CN107204006A (en) * | 2017-06-01 | 2017-09-26 | 大连海事大学 | A kind of static target detection method based on double background difference |
-
2018
- 2018-10-18 CN CN201811217578.7A patent/CN109492543A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1514408A (en) * | 2002-09-12 | 2004-07-21 | 上海交通大学 | Infra red detecting and tracing method for weak target under complex background condition |
| US8724850B1 (en) * | 2011-06-21 | 2014-05-13 | The United States Of America As Represented By The Secretary Of The Navy | Small object detection using meaningful features and generalized histograms |
| CN103310419A (en) * | 2013-06-18 | 2013-09-18 | 深圳大学 | Small target image processing device |
| CN107204006A (en) * | 2017-06-01 | 2017-09-26 | 大连海事大学 | A kind of static target detection method based on double background difference |
Non-Patent Citations (2)
| Title |
|---|
| 崔璇: "天空背景下红外小目标检测算法研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
| 崔璇等: "一种有效的红外小目标检测方法", 《光子学报》 * |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110111261A (en) * | 2019-03-28 | 2019-08-09 | 福州瑞芯微电子股份有限公司 | Adaptive equalization processing method, electronic equipment and the computer readable storage medium of image |
| CN110135344A (en) * | 2019-05-16 | 2019-08-16 | 电子科技大学 | Based on the method for detecting infrared puniness target for weighting fixed order expression |
| CN110276788A (en) * | 2019-06-12 | 2019-09-24 | 北京轩宇空间科技有限公司 | Method and apparatus for infrared imaging formula target seeker target following |
| CN110400294A (en) * | 2019-07-18 | 2019-11-01 | 湖南宏动光电有限公司 | A kind of Infrared Target Detection system and detection method |
| CN110400294B (en) * | 2019-07-18 | 2023-02-07 | 湖南宏动光电有限公司 | Infrared target detection system and detection method |
| CN111079546A (en) * | 2019-11-22 | 2020-04-28 | 重庆师范大学 | Unmanned aerial vehicle pest detection method |
| CN111428573B (en) * | 2020-03-02 | 2021-06-25 | 南京莱斯电子设备有限公司 | Infrared weak and small target detection false alarm suppression method under complex background |
| CN111428573A (en) * | 2020-03-02 | 2020-07-17 | 南京莱斯电子设备有限公司 | Infrared weak and small target detection false alarm suppression method under complex background |
| CN111337133B (en) * | 2020-03-02 | 2021-08-03 | 浙江大华技术股份有限公司 | Infrared data generation method and device and infrared data analysis method and device |
| CN111337133A (en) * | 2020-03-02 | 2020-06-26 | 浙江大华技术股份有限公司 | Infrared data generation method and device and infrared data analysis method and device |
| CN111401195A (en) * | 2020-03-10 | 2020-07-10 | 上海航天控制技术研究所 | Sea surface target detection method based on multiband infrared image |
| CN111401195B (en) * | 2020-03-10 | 2023-11-14 | 上海航天控制技术研究所 | Sea surface target detection method based on multiband infrared image |
| CN112954138A (en) * | 2021-02-20 | 2021-06-11 | 东营市阔海水产科技有限公司 | Aquatic economic animal image acquisition method, terminal equipment and movable material platform |
| CN113613007A (en) * | 2021-07-19 | 2021-11-05 | 青岛信芯微电子科技股份有限公司 | Three-dimensional color lookup table generation method and display device |
| CN113613007B (en) * | 2021-07-19 | 2024-03-05 | 青岛信芯微电子科技股份有限公司 | Three-dimensional color lookup table generation method and display device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109492543A (en) | The small target detecting method and system of infrared image | |
| Nandi | Detection of human brain tumour using MRI image segmentation and morphological operators | |
| CN107808383B (en) | Rapid detection method for SAR image target under strong sea clutter | |
| CN103927751A (en) | Water surface optical visual image target area detection method based on gradient information fusion | |
| Kaspers | Blob detection | |
| CN107705313B (en) | Remote sensing image ship target segmentation method | |
| Wang et al. | Clutter-adaptive infrared small target detection in infrared maritime scenarios | |
| Wang et al. | Extracting oil slick features from VIIRS nighttime imagery using a Gaussian filter and morphological constraints | |
| Bhadauria et al. | Comparison between various edge detection methods on satellite image | |
| AlAzawee et al. | Using morphological operations—Erosion based algorithm for edge detection | |
| CN104899842A (en) | Ranked-order adaptively extremum median filtering method for remote line structured light image | |
| Pujar et al. | Medical image segmentation based on vigorous smoothing and edge detection ideology | |
| Haider et al. | A hybrid method for edge continuity based on Pixel Neighbors Pattern Analysis (PNPA) for remote sensing satellite images | |
| Rana et al. | Review on traditional methods of edge detection to morphological based techniques | |
| Kumar et al. | Edge detection of angiogram images using the classical image processing techniques | |
| Liu et al. | Automatic detection of ship target and motion direction in visual images | |
| Jadhav et al. | FPGA based object tracking system | |
| Hsieh et al. | A novel approach to the detection of small objects with low contrast | |
| Chen et al. | Level sets with self-guided filtering for marine oil spill segmentation | |
| Nigam et al. | Evaluation and analysis of different type of edge detection techniques on cartridge case image | |
| Swathika et al. | Edge Detection Using Simple Image Arithmetic | |
| Wang et al. | An automatic bridge detection technique for high resolution SAR images | |
| Ihmeida et al. | Deep despeckling of SAR images to improve change detection performance | |
| Karim et al. | Comparing the Main Approaches of Image Segmentation | |
| Nosrati et al. | A method for detection and extraction of circular shapes from noisy images using median filter and CHT |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190319 |
|
| RJ01 | Rejection of invention patent application after publication |