CN104517267B - Infrared image enhancement and reestablishment method based on spectra inversion - Google Patents
Infrared image enhancement and reestablishment method based on spectra inversion Download PDFInfo
- Publication number
- CN104517267B CN104517267B CN201410811549.9A CN201410811549A CN104517267B CN 104517267 B CN104517267 B CN 104517267B CN 201410811549 A CN201410811549 A CN 201410811549A CN 104517267 B CN104517267 B CN 104517267B
- Authority
- CN
- China
- Prior art keywords
- infrared image
- infrared
- image
- sub
- spectrum
- 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.)
- Expired - Fee Related
Links
Landscapes
- Image Processing (AREA)
Abstract
The invention discloses an infrared image enhancement and reestablishment method based on spectra inversion, belongs to the field of infrared image processing, and particularly relates to a new infrared image enhancement and detail reestablishment method and a new computational imaging method. The method uses a seismic trace high-resolution reestablishment thought in seismic signal processing proposed recently for reference to carry out the spectra inversion (reflected energy reconstruction) of infrared images. The method carries out the spectra inversion in dominant frequency bands by establishing an infrared reflected energy model and reflected energy odd-even decomposition model, and finally realizes the improvement of infrared image detail resolution ratio. The method can highlight the reflected energy details of weak and small targets on the basis of without increasing the image space resolution, thus improving the target contrast ratio and the energy feature. The method is beneficial for subsequent detection and tracking of the infrared weak and small targets, and has a wider application prospect.
Description
Technical field
The invention belongs to infrared image processing field, more particularly to a kind of new infrared image enhancement and details method for reconstructing
With new calculating imaging method.The method uses for reference seismic channel super-resolution reconstruction thought in the seismic data processing for proposing in the recent period,
Carry out " spectrum " inverting (reflected energy reconstruct) of infrared image.
Background technology
Infrared imagery technique is widely used in fields such as Modern Traffic, safety monitorings, in IRDS
In, infrared target detection and tracking are its key technology and nucleus module.Because remote infrared imaging often brings image pair
The defects such as when signal to noise ratio is low, imaging area is little for, noise lower than degree, amorphism and textural characteristics, need infrared to what is collected
Image is pre-processed, and improves the detail resolution of image, suppressing background energy, improving target area energy and projecting target
Details, is conducive to the detect and track of follow-up infrared small object.
Traditional super-resolution image reconstruction (Super resolution image re-construction, SRIR/
SR the method with signal transacting and image procossing) is referred to, by existing low resolution (Low- by way of software algorithm
Resolution, LR) image is converted into the technology of high-resolution (High-resolution, HR) image, and main method is divided into base
In the super-resolution rebuilding of Image Reconstruction, and the super-resolution rebuilding based on study etc..The technology is by using one group of low-quality
Amount, low-resolution image (or motion sequence) are producing single width high-quality, high-definition picture.Conventional method often brings image
The change of bulk, and can not targetedly strengthen the energy feature of area-of-interest.
In seismic prospecting signal transacting, the inversion method that inverting is a kind of fine identification thin layer and reflectance factor is composed.Spectrum
Inverting is set up on the basis of seismic signal spectral factorization, and by spectral factorization, we extract seismic channel instantaneous attribute, these category
Property can be used to properly increase the resolution ratio of seismic data.In frequency domain, the estimation and reconstruct of reflectance factor is carried out, to enter one
Step improves the resolution ratio of inversion result, as composes inverting.
1980s, the concept of spectral factorization is suggested.The complex trace analysis that Taner et al. (1979) is proposed defines earliest
Theory of spectrum dissolution basis.As uncertainty principle is introduced into digital processing field, Spectral Decomposition Technique starts to develop rapidly.Borrow
With the method for different time-frequency basic function signal Analysis in mirror Digital Signal Processing, physical prospecting scholars are in succession by Gabor transformation, company
The methods such as continuous wavelet transformation, wavelet transform, S-transformation introduce seismic data analysis.Popularization along with these new algorithms should
With the extraction result of seismic data instantaneous attribute there has also been significant improvement.Afterwards, Portniaguine (2004) proposes earthquake
Signal inverse spectral decomposition method, Sinha, Castagna et al. (2005) propose is used for earthquake spectral factorization by continuous wavelet transform
Method etc., the proposition of these methods promotes the development of seismic data Spectral Decomposition Technique.
Spectrum inverting is spectral factorization and the combination of inversion algorithm.Through years of researches and development, spectrum inversion method is proved to
For a kind of effective, method that seismic data image thin layer resolution ratio can be improved.In not noise and known seismic wavelet
Ideally, the method can recognize in theory thin layer of the thickness less than tuning thickness, and accurately can depict ground
The border of layer.Spectrum inverting object function there is preferable convergence and restriction ability, by adjust reflectance factor position and
Size, under the constraint of frequency domain object function, is obtained high-resolution reflection coefficient sequence.
Although spectrum inversion method is from the development for proposing to experienced till now more than ten years, in key technology and its application,
Also many problems need to go further investigation, it may be said that it or a kind of new method and technology.The developing history of spectrum inverting is looked back, can
To be divided into following several stages：
Partyka (1999) etc. has found that spectral factorization method is applied in seismic data interpretation, can properly increase seismic data
Resolution ratio；2004, Spectral Decomposition Technique was combined seismic inversion by Partyka etc., proposed spectrum inversion theory.Satinder
Chopra, JohnP.Castagna etc. (2006) carry out inverting with the local spectrum data that spectral factorization is obtained, and calculate thin layer anti-
Coefficient is penetrated, the resolution ratio of inversion result is better than conventional inverting, and it is false without the need for prior model, reflectance factor to indicate that the method has
If, layer position constraint, Log-constrained the advantages of.Charles I.Puryear and John P.Castagna (2008) is reflectance factor
Even component and odd component are resolved into, the fundamental formular of spectrum inverting is derived, and the method is demonstrated by modeling computation be can be used to
Differentiate thin layer of the underground less than tuning thickness；Sanyi Yuan etc. (2009) are theoretically discussed and compose at this stage what inverting was present
Problem and its applicable elements, and the ill-posedness to composing inverting has carried out detailed analysis.Satinder.Chopra and John
P.Castagna etc. (2009) is applied to real data by inversion method is composed, and has calculated sparse reflectance factor, and this is not only
The feasibility of spectrum inversion method is demonstrated, the application prospect and future thrust of spectrum inverting is also indicated that.
The content of the invention
Present invention design is a kind of based on wavelet decomposition and to use for reference the image processing method of seismic prospecting signal spectrum inversion theory.
By setting up infrared image reflected energy model, " spectrum " inversion method is adopted, in the case where spatial resolution is not changed, realized
The detail resolution of image strengthens.The method for Infrared Image Information subsequent treatment, such as the identification of Weak target, background
Classification etc. provides more useful and efficient information, can improve infrared target detection and discrimination.
A kind of infrared image enhancement and method for reconstructing based on " spectrum " inverting of the invention, the method includes：
Step 1：Obtain a pending infrared image；
Step 2：Infrared image is built using small echo and decomposes dictionary, the atom in dictionary is sparse base or orthogonal basis；
Step 3：The imagery exploitation dictionary that step 1 is obtained is decomposed, and obtains some subbands, and obtain each subband be
Number；
Step 4：Set up each sub-belt energy model；
Step 5：Each sub-band coefficients are carried out with odd even component decomposition, odd component coefficient and even component coefficient is obtained, it is possible to
Odd even component subband and its corresponding energy model are constructed respectively；
Step 6：Set up infrared image detail resolution evaluation criterion；
Step 7：The odd component coefficient and even component coefficient of each subband that step 5 is obtained is multiplied by different transformation factors,
Change its size, then using the coefficient reconstruct infrared image after changing, judged after reduction by the evaluation criterion that step 6 is set up
Whether image resolution ratio reaches requirement；
Step 8：The infrared image collected under several similar situations is carried out according to step 1 to step 7 identical method
Details is strengthened, and obtains the component coefficient transformation factor of different images, and further according to the component coefficient transformation factor for obtaining one is obtained
The component coefficient transformation factor for commonly using, the details reinforcing of the infrared image for obtaining under similar situation.
Dictionary is built using orthogonal wavelet in wherein described step 2.
Dictionary is utilized to be high-frequency sub-band and low frequency sub-band to obtaining picture breakdown in wherein described step 3, and it is right to obtain its
The sub-band coefficients answered；
Wherein described step 6 sets up infrared image detail resolution evaluation criterion：C=Cdef×CSNR, wherein：CdefFor
The definition of image, CSNRFor the signal to noise ratio of image；
Wherein described step 7 is concretely comprised the following steps：
Step 7-1：Set up forward model
M is the optimum sub-band coefficients of requirement, K1And K2For inverse transformation operator, H1And H2For the degeneracy operator of image, for example by
In the fuzzy, ambient noise caused by camera imaging quality, atmospheric perturbation etc., F represents Image Sub-Band, and n is one and small disturbs
It is dynamic, represent the otherness of forward model and actual signal；
Step 7-2：Object function is obtained according to step 7-1：
Wherein δ represents infrared image detail resolution Evaluation threshold, and α is regularization factors.
A kind of infrared image enhancement and method for reconstructing based on " spectrum " inverting of the invention, belongs to infrared image processing field,
More particularly to a kind of new infrared image enhancement and details method for reconstructing and new calculating imaging method.The method is red by setting up
External reflectance energy model and the odd, even decomposition model of reflected energy, carry out the spectrum inverting in dominant frequency band, finally realize infrared image
The raising of detail resolution.The invention can project the " anti-of Weak target on the basis of image spatial resolution is not increased
Penetrate " energy details, improves target contrast and energy feature.Be conducive to the detect and track of follow-up infrared small object, have
Larger application prospect.
Description of the drawings
Fig. 1 is the odd even component decomposing schematic representation of the signal containing two pulses；
Fig. 2 is the practical IR image that a width contains Weak target；
Fig. 3 is the result figure that infrared image different frequency bands subgraph carries out odd even component decomposition；
Fig. 4 is to adopt the method proposed in the present invention, the infrared image for obtaining " spectrum " inversion result figure；
Fig. 5 is the energy profile of infrared image " spectrum " inversion result, and encircled portion is represented and obtains enhanced infrared small and weak
Target energy；
Fig. 6 is method flow diagram.
Specific embodiment
The present invention is specifically described below in conjunction with the accompanying drawings.
Step 1：One pending infrared image f (x, y) of input, its size is M × N；
Step 2：Build infrared image and decompose dictionary, the atom in dictionary is sparse base or orthogonal basis.In the present invention, adopt
Dictionary is built with orthogonal wavelet (such as Haar small echos)；
Step 3：The image of reading is carried out into j multi-scale wavelet decomposition：
In wavelet decomposition, high frequency coefficient is calculated by following formula：
And low frequency coefficient
Step 4：According to the result of step 3, infrared image low-and high-frequency reflected energy model (sub-belt energy model) is built.By
Wavelet sub-band energy model, high-frequency energy is adopted to be expressed as in here
Here j represents wavelet decomposition series.
Low frequency energy is expressed as
Here j0Represent first order wavelet decomposition series.
Step 5：High-frequency sub-band coefficient is chosen, odd even component decomposition is carried out.Here by taking the signal containing two pulses as an example,
Illustrate the Parity-decomposition principle (see Fig. 1) of signal.In the spectrum inverting of seismic signal, it is believed that the even component of reflectance factor can
To improve thin layer resolution ratio, and odd component weakens thin layer resolution ratio.For the infrared image containing Weak target (see Fig. 2), I
Its high-frequency sub-band coefficient is carried out into Parity-decomposition, obtain odd component sub-band coefficients and even component sub-band coefficients, it is possible to synthesize
Odd even component high frequency subgraph (see Fig. 3)；
Step 6：Setting up infrared image detail resolution strengthens evaluation criterion.For the characteristics of infrared image and follow-up mesh
The needs of mark detection, take definition and signal to noise ratio as evaluation criterion.The definition of image is expressed as
Wherein
△fx(i, j)=f (i, j)-f (i-1, j)
(7)
△fy(i, j)=f (i, j)-f (i, j-1)
Signal to noise ratio is expressed as
Wherein μtRepresent the average pixel value of target area, μbAnd σbRespectively represent target surrounding pixel point mean value and
Standard deviation.Consider both the above evaluation criterion, the constraints set up in this problem
Wherein β is a constant, can be adjusted according to actual needs.
Step 7：Derived object function, solution can strengthen " puppet " the subband system of infrared small object imaging plane energy
Number.Image containing infrared small object can be considered as being superimposed for projection of the target area on imaging plane and background area,
Forward model is set up accordingly
M is amount to be asked, i.e., the optimum sub-band coefficients required by us, K1And K2As corresponding to conversion step (3) Suo Shi
Inverse transformation operator, H1And H2Represent the degeneracy operator of image, for example due to the fuzzy, ambient noise caused by camera imaging quality,
Atmospheric perturbation etc., F represents the subband for treating inverting.Solve following object function：
Wherein δ represents that infrared image detail resolution obtains enhanced degree, and concrete numerical value can determine according to actual needs.
Said process is infrared image " spectrum " refutation process, can improve " puppet " reflected energy on Weak target imaging plane；Instead
The method of drilling can adopt nonlinear optimization algorithm, such as simulated annealing, random hill-climbing algorithm, and trying to achieve can strengthen infrared small and weak mesh
The optimum sub-belt energy coefficient of mark detail resolution." puppet " sub-band coefficients tried to achieve are utilized, new high and low frequency can be reconstructed
Subgraph；
Step 8：Using the result of step 7, Image Reconstruction is carried out, obtain enhancing the infrared image of detail resolution, it is defeated
Go out result (see Fig. 4, Fig. 5).Encircled portion is represented and has obtained enhanced infrared small object energy in Fig. 5.
Claims (4)
1. one kind is based on the infrared image enhancement and method for reconstructing of " spectrum " inverting, and the method includes：
Step 1：Obtain a pending infrared image；
Step 2：Infrared image is built using small echo and decomposes dictionary, the atom in dictionary is sparse base or orthogonal basis；
Step 3：The imagery exploitation dictionary that step 1 is obtained is decomposed, and obtains some subbands, and obtains the coefficient of each subband；
Step 4：Set up each sub-belt energy model；
Step 5：Each sub-band coefficients are carried out with odd even component decomposition, odd component coefficient and even component coefficient is obtained, it is possible to respectively
Construct odd even component subband and its corresponding energy model；
Step 6：Set up infrared image detail resolution evaluation criterion；
Step 7：Derived object function, solution can strengthen " puppet " sub-band coefficients of infrared small object imaging plane energy；
The step 7 is concretely comprised the following steps：
Step 7-1：Set up forward model
M is the optimum sub-band coefficients of requirement, K1And K2For inverse transformation operator, H1And H2For the degeneracy operator of image, due to camera into
Fuzzy, ambient noise, atmospheric perturbation as caused by quality is caused, and F represents Image Sub-Band, and n is a small disturbance, represents
The otherness of forward model and actual signal；
Step 7-2：Object function is obtained according to step 7-1：
Wherein δ represents infrared image detail resolution Evaluation threshold, and α is regularization factors；
Step 8：Using the result of step 7, Image Reconstruction is carried out, obtain enhancing the infrared image of detail resolution, output knot
Really.
2. as claimed in claim 1 one kind is based on the infrared image enhancement and method for reconstructing of " spectrum " inverting, it is characterised in that
Dictionary is built using orthogonal wavelet in the step 2.
3. as claimed in claim 1 one kind is based on the infrared image enhancement and method for reconstructing of " spectrum " inverting, it is characterised in that
Utilize dictionary to be high-frequency sub-band and low frequency sub-band to obtaining picture breakdown in the step 3, and obtain its corresponding sub-band coefficients.
4. as claimed in claim 1 one kind is based on the infrared image enhancement and method for reconstructing of " spectrum " inverting, it is characterised in that
The step 6 sets up infrared image detail resolution evaluation criterion：C=Cdef×CSNR, wherein：CdefFor the definition of image,
CSNRFor the signal to noise ratio of image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410811549.9A CN104517267B (en) | 2014-12-23 | 2014-12-23 | Infrared image enhancement and reestablishment method based on spectra inversion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410811549.9A CN104517267B (en) | 2014-12-23 | 2014-12-23 | Infrared image enhancement and reestablishment method based on spectra inversion |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104517267A CN104517267A (en) | 2015-04-15 |
| CN104517267B true CN104517267B (en) | 2017-05-10 |
Family
ID=52792517
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410811549.9A Expired - Fee Related CN104517267B (en) | 2014-12-23 | 2014-12-23 | Infrared image enhancement and reestablishment method based on spectra inversion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104517267B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107193046A (en) * | 2016-03-14 | 2017-09-22 | 中国石油化工股份有限公司 | A kind of sand thickness Forecasting Methodology and system based on spectrum inverting |
| CN107065006B (en) * | 2017-01-23 | 2019-06-11 | 武汉大学 | A kind of seismic signal coding method based on online dictionary updating |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2009008778A1 (en) * | 2007-07-09 | 2009-01-15 | Flir Systems Ab | Method of processing an infrared image, infrared image capturing system and computer readable medium |
| CN101833752A (en) * | 2010-04-20 | 2010-09-15 | 南京航空航天大学 | Pretreatment method for decomposed and reconstituted infrared small targets based on singular values |
| EP2249705A2 (en) * | 2008-01-30 | 2010-11-17 | The Regents of the University of California | Dose reduction and image enhancement in tomography through the utilization of the object's surroundings as dynamic constraints |
| CN102222322A (en) * | 2011-06-02 | 2011-10-19 | 西安电子科技大学 | Multiscale non-local mean-based method for inhibiting infrared image backgrounds |
| CN102800059A (en) * | 2012-07-05 | 2012-11-28 | 清华大学 | Image visibility enhancing method with assistance of near-infrared image |
| CN103247069A (en) * | 2013-05-16 | 2013-08-14 | 中国电子科技集团公司第四十一研究所 | Ultraviolet scene simulation method and system based on radiating energy characteristics and spectrum characteristics |
| CN103413289A (en) * | 2013-08-30 | 2013-11-27 | 中国神华能源股份有限公司 | Image processing method and device for mining area |
| CN103927752A (en) * | 2014-04-21 | 2014-07-16 | 国家电网公司 | Method for monitoring operating state of electrical equipment |
| CN104036455A (en) * | 2013-03-06 | 2014-09-10 | 江苏理工学院 | Infrared image detail enhancement method based on second-generation wavelet |
-
2014
- 2014-12-23 CN CN201410811549.9A patent/CN104517267B/en not_active Expired - Fee Related
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2009008778A1 (en) * | 2007-07-09 | 2009-01-15 | Flir Systems Ab | Method of processing an infrared image, infrared image capturing system and computer readable medium |
| EP2249705A2 (en) * | 2008-01-30 | 2010-11-17 | The Regents of the University of California | Dose reduction and image enhancement in tomography through the utilization of the object's surroundings as dynamic constraints |
| CN101833752A (en) * | 2010-04-20 | 2010-09-15 | 南京航空航天大学 | Pretreatment method for decomposed and reconstituted infrared small targets based on singular values |
| CN102222322A (en) * | 2011-06-02 | 2011-10-19 | 西安电子科技大学 | Multiscale non-local mean-based method for inhibiting infrared image backgrounds |
| CN102800059A (en) * | 2012-07-05 | 2012-11-28 | 清华大学 | Image visibility enhancing method with assistance of near-infrared image |
| CN104036455A (en) * | 2013-03-06 | 2014-09-10 | 江苏理工学院 | Infrared image detail enhancement method based on second-generation wavelet |
| CN103247069A (en) * | 2013-05-16 | 2013-08-14 | 中国电子科技集团公司第四十一研究所 | Ultraviolet scene simulation method and system based on radiating energy characteristics and spectrum characteristics |
| CN103413289A (en) * | 2013-08-30 | 2013-11-27 | 中国神华能源股份有限公司 | Image processing method and device for mining area |
| CN103927752A (en) * | 2014-04-21 | 2014-07-16 | 国家电网公司 | Method for monitoring operating state of electrical equipment |
Non-Patent Citations (2)
| Title |
|---|
| "叠后地震数据的谱反演处理技术及其应用浅析";曹鉴华 等;《地球物理学进展》;20130228;第28卷(第1期);第387-392页 * |
| "谱反演在地震属性解释中的应用";刘万金 等;《石油地球物理勘探》;20130630;第48卷(第3期);第424-428页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104517267A (en) | 2015-04-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103873743B (en) | Video de-noising method based on structure tensor and Kalman filtering | |
| CN103093433B (en) | Natural image denoising method based on regionalism and dictionary learning | |
| CN109658351B (en) | Hyperspectral image denoising method combining L0 gradient constraint and local low-rank matrix recovery | |
| CN104408700A (en) | Morphology and PCA (principal component analysis) based contourlet fusion method for infrared and visible light images | |
| CN110826389B (en) | Gait recognition method based on attention 3D frequency convolution neural network | |
| CN101930604B (en) | Infusion method of full-color image and multi-spectral image based on low-frequency correlation analysis | |
| CN105279740A (en) | Image denoising method based on sparse regularization | |
| CN104008537A (en) | Novel noise image fusion method based on CS-CT-CHMM | |
| CN102306381B (en) | Method for fusing images based on beamlet and wavelet transform | |
| CN104063886A (en) | Nuclear magnetic resonance image reconstruction method based on sparse representation and non-local similarity | |
| CN103400383A (en) | SAR (synthetic aperture radar) image change detection method based on NSCT (non-subsampled contourlet transform) and compressed projection | |
| CN104252704A (en) | Total generalized variation-based infrared image multi-sensor super-resolution reconstruction method | |
| CN103514600B (en) | A kind of infrared target fast robust tracking based on rarefaction representation | |
| CN103971354A (en) | Method for reconstructing low-resolution infrared image into high-resolution infrared image | |
| CN104917933B (en) | The low-light video denoising method estimated based on phase equalization and scene motion | |
| CN104778670A (en) | Fractal-wavelet self-adaption image denoising method based on multivariate statistical model | |
| CN104517267B (en) | Infrared image enhancement and reestablishment method based on spectra inversion | |
| CN105913402A (en) | Multi-remote sensing image fusion denoising method based on DS evidence theory | |
| CN105809650A (en) | Bidirectional iteration optimization based image integrating method | |
| CN102314675B (en) | Wavelet high-frequency-based Bayesian denoising method | |
| Lu et al. | Structure-texture parallel embedding for remote sensing image super-resolution | |
| Thakur et al. | Gradient and multi scale feature inspired deep blind gaussian denoiser | |
| Shang et al. | Seismic random noise suppression using an adaptive nonlocal means algorithm | |
| CN105069762A (en) | Image denoising method based on Shearlet transform and non-linear diffusion | |
| Zhou et al. | Deep learning with fault prior for 3-D seismic data super-resolution |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20170510 Termination date: 20191223 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |