CN111667517A - Infrared polarization information fusion method and device based on wavelet packet transformation - Google Patents
Infrared polarization information fusion method and device based on wavelet packet transformation Download PDFInfo
- Publication number
- CN111667517A CN111667517A CN202010507681.6A CN202010507681A CN111667517A CN 111667517 A CN111667517 A CN 111667517A CN 202010507681 A CN202010507681 A CN 202010507681A CN 111667517 A CN111667517 A CN 111667517A
- Authority
- CN
- China
- Prior art keywords
- polarization information
- infrared polarization
- image
- wave
- images
- 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
- 230000010287 polarization Effects 0.000 title claims abstract description 300
- 230000009466 transformation Effects 0.000 title claims abstract description 44
- 238000007500 overflow downdraw method Methods 0.000 title claims abstract description 29
- 238000000034 method Methods 0.000 claims abstract description 22
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 20
- 230000004927 fusion Effects 0.000 claims description 61
- 239000011159 matrix material Substances 0.000 claims description 53
- 238000003384 imaging method Methods 0.000 claims description 11
- 238000004590 computer program Methods 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000000605 extraction Methods 0.000 claims description 5
- 238000013519 translation Methods 0.000 claims description 5
- 238000005259 measurement Methods 0.000 claims description 4
- 238000001514 detection method Methods 0.000 abstract description 13
- 238000011426 transformation method Methods 0.000 abstract description 2
- 230000005855 radiation Effects 0.000 description 29
- 239000013077 target material Substances 0.000 description 4
- 238000012360 testing method Methods 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 3
- 230000005457 Black-body radiation Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000007499 fusion processing Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 238000004873 anchoring Methods 0.000 description 1
- 230000002457 bidirectional effect Effects 0.000 description 1
- 238000005315 distribution function Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/10—Image acquisition
- G06V10/12—Details of acquisition arrangements; Constructional details thereof
- G06V10/14—Optical characteristics of the device performing the acquisition or on the illumination arrangements
- G06V10/143—Sensing or illuminating at different wavelengths
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10048—Infrared image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20048—Transform domain processing
- G06T2207/20064—Wavelet transform [DWT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20212—Image combination
- G06T2207/20221—Image fusion; Image merging
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Image Processing (AREA)
Abstract
The invention relates to an infrared polarization information fusion method, device, equipment, system and computer storage medium based on wavelet packet transformation, wherein the method comprises the following steps: acquiring infrared polarization information images of different wave bands and registering; the infrared polarization information images of different wave bands comprise long wave infrared polarization information images and medium wave infrared polarization information images; performing wavelet packet decomposition on the registered infrared polarization information images of different wave bands, fusing corresponding frequency band sub-images obtained by decomposing the medium wave infrared polarization information images and the long wave infrared polarization information images, and reconstructing through wavelet packets to obtain fused images; and carrying out target identification based on the fused image. According to the invention, aiming at the characteristic difference of the infrared polarization information of the target surface in the medium and long wave bands, the medium and long wave band infrared polarization degree images are fused based on a wavelet packet transformation method, and the contour characteristics such as high-frequency details and the like of the fused images are prominent, so that the accuracy of target detection and identification can be improved.
Description
Technical Field
The invention relates to the field of polarization information image fusion, in particular to an infrared polarization information fusion method, device, equipment and system based on wavelet packet transformation and a computer readable storage medium.
Background
The infrared polarization image mainly improves the target detection capability through the detail characteristics of the contour, the edge and the like of the target. Theoretical analysis and experiments show that the infrared polarization characteristic of the target surface is related to the detection wavelength and is not a fixed value. Typical target surface red polarization characteristics are generally not the same in the mid and long wavelength bands, and because mid-band infrared polarization information characteristics are more affected by ambient radiation, the target surface mid-wave infrared polarization characteristics are generally not significant. In view of the difference of the infrared polarization characteristics of the target surface in the medium and long wave bands, in order to improve the identification accuracy of the polarization characteristics based on the target edge profile, a medium and long wave infrared polarization information fusion method needs to be researched.
According to the characteristics of the infrared polarization information image, the fused image not only needs to keep the low-frequency characteristics of the infrared image, but also needs to highlight detail information (mostly expressed as high-frequency characteristics) such as scene edges, outlines and the like in the polarization image so as to improve the target detection and identification capability of the fused image. Therefore, it is desirable to provide an infrared polarization information fusion method, which can effectively improve the accuracy of target identification and detection based on the infrared polarization profile characteristics of the target surface.
Disclosure of Invention
The technical problem to be solved by the present invention is to provide an infrared polarization information fusion method, apparatus, device, system and computer readable storage medium based on wavelet packet transformation, which improves the accuracy of target identification and detection based on the infrared polarization profile characteristics of the target surface.
In order to solve the above technical problem, a first aspect of the present invention provides an infrared polarization information fusion method based on wavelet packet transformation, including the following steps:
acquiring infrared polarization information images of different wave bands and registering; the infrared polarization information images of different wave bands comprise long wave infrared polarization information images and medium wave infrared polarization information images;
performing wavelet packet decomposition on the registered infrared polarization information images of different wave bands, fusing corresponding frequency band sub-images obtained by decomposing the medium wave infrared polarization information image and the long wave infrared polarization information image, and reconstructing through wavelet packets to obtain a fused image;
and carrying out target identification based on the fused image.
In the method for fusing infrared polarization information based on wavelet packet transformation according to the present invention, preferably, the fusing the corresponding band subimages obtained by decomposing the medium wave infrared polarization information image and the long wave infrared polarization information image includes performing low-frequency component fusion by using a weighted average method:
WPl,F＝aWPl,A+bWPl,B
wherein, WPi,FFor fusing the low frequency component part of the image, a and b are weighting coefficients, and a + b is 1, WPi,AFor low-frequency subimages, WP, obtained after decomposition of the mid-wave infrared polarization information imagei,BThe long-wave infrared polarization information image is decomposed to obtain a low-frequency sub-image.
In the method for fusing infrared polarization information based on wavelet packet transformation according to the present invention, preferably, the fusing the corresponding band subimages obtained by decomposing the medium wave infrared polarization information image and the long wave infrared polarization information image includes performing high frequency component fusion by using a fusion method based on local region features: calculating the variance matching degree of the long wave infrared polarization information image and the medium wave infrared polarization information image with the local area corresponding to the kth node coefficient matrix of the 1 st layerM1k,AB(m,n)：
Wherein, WP1k,A(m, n) is the k node coefficient matrix term of the 1 st layer decomposed by the medium wave infrared polarization information image, WP1k,B(m, n) is the k node coefficient matrix item of the 1 st layer decomposed for the long wave infrared polarization information image,
the average value of a Q area with a point (m, n) as the center is used as a k node coefficient matrix of a layer 1 decomposed by the medium wave infrared polarization information image;
the average value of a local area Q taking a point (m, n) as the center of a kth node coefficient matrix of a layer 1 decomposed on a long-wave infrared polarization information image; g1k,A(m, n) is the local region variance with the point (m, n) as the center in the kth node coefficient matrix of the layer 1 of the medium-wave infrared polarization information image decomposition; g1k,B(m, n) is the local area variance with the point (m, n) as the center in the kth node coefficient matrix of the layer 1 decomposed by the long-wave infrared polarization information image;
judging variance matching degree M1k,ABIf (m, n) is less than the preset threshold, if yes, the k node coefficient matrix item WP of the 1 st layer of the high-frequency component part of the fused image1k,F(m, n) is:
otherwise, fusing the kth node coefficient matrix term WP of the layer 1 of the high-frequency component part of the image1k,F(m, n) is:
wherein, the weighting coefficient of the kth node of the 1 st layer
In the method for fusing infrared polarization information based on wavelet packet transformation according to the present invention, preferably, the acquiring and registering infrared polarization information images of different wavebands includes:
performing edge extraction on the infrared polarization information image, and respectively extracting edge linear characteristics of targets in the long-wave infrared polarization information image and the medium-wave infrared polarization information image;
taking the infrared polarization information image of the current wave band as a reference, solving the deflection angle and the two-dimensional scaling of the infrared polarization information image of the other wave band, and correcting the long-wave infrared polarization information image of the other wave band according to the deflection angle and the two-dimensional scaling;
and performing autocorrelation calculation based on the extracted edge linear features, determining the two-dimensional offset of the infrared polarization information image of the current waveband according to the peak occurrence position, and performing translation adjustment on the infrared polarization information image of the other waveband according to the two-dimensional offset to realize registration of the infrared polarization information images of different wavebands.
In a second aspect of the present invention, an infrared polarization information fusion apparatus based on wavelet packet transformation is provided, including:
the image registration unit is used for acquiring infrared polarization information images of different wave bands and registering the infrared polarization information images; the infrared polarization information images of different wave bands comprise long wave infrared polarization information images and medium wave infrared polarization information images;
the image fusion unit is used for performing wavelet packet decomposition on the registered infrared polarization information images of different wave bands, fusing corresponding frequency band sub-images obtained by decomposing the medium wave infrared polarization information images and the long wave infrared polarization information images, and reconstructing through wavelet packets to obtain fusion images;
and the target identification unit is used for carrying out target identification based on the fused image.
In the infrared polarization information fusion device based on wavelet packet transformation according to the present invention, preferably, the image fusion unit performs low frequency component fusion by means of weighted average:
WPl,F＝aWPl,A+bWPl,B
wherein, WPi,FFor fusing the low frequency component part of the image, a and b are weighting coefficients, and a + b is 1, WPi,AFor low-frequency subimages, WP, obtained after decomposition of the mid-wave infrared polarization information imagei,BThe long-wave infrared polarization information image is decomposed to obtain a low-frequency sub-image.
In the infrared polarization information fusion device based on wavelet packet transformation according to the present invention, preferably, the image fusion unit performs high frequency component fusion by using a fusion method based on local region features: calculating the variance matching degree M of the long wave infrared polarization information image and the medium wave infrared polarization information image and the local area corresponding to the kth node coefficient matrix of the 1 st layer1k,AB(m,n)：
Wherein, WP1k,A(m, n) is the k node coefficient matrix term of the 1 st layer decomposed by the medium wave infrared polarization information image, WP1k,B(m, n) is the k node coefficient matrix item of the 1 st layer decomposed for the long wave infrared polarization information image,
the average value of a Q area with a point (m, n) as the center is used as a k node coefficient matrix of a layer 1 decomposed by the medium wave infrared polarization information image;
the average value of a local area Q taking a point (m, n) as the center of a kth node coefficient matrix of a layer 1 decomposed on a long-wave infrared polarization information image; g1k,A(m, n) is the local region variance with the point (m, n) as the center in the kth node coefficient matrix of the layer 1 of the medium-wave infrared polarization information image decomposition; g1k,B(m, n) is the kth node system of the 1 st layer of long-wave infrared polarization information image decompositionLocal area variance centered at point (m, n) in the number matrix;
judging variance matching degree M1k,ABIf (m, n) is less than the preset threshold, if yes, the k node coefficient matrix item WP of the 1 st layer of the high-frequency component part of the fused image1k,F(m, n) is:
otherwise, fusing the kth node coefficient matrix term WP of the layer 1 of the high-frequency component part of the image1k,F(m, n) is:
wherein, the weighting coefficient of the kth node of the 1 st layer
In the infrared polarization information fusion device based on the wavelet packet transformation according to the present invention, preferably, the image registration unit performs registration by:
performing edge extraction on the infrared polarization information image, and respectively extracting edge linear characteristics of targets in the long-wave infrared polarization information image and the medium-wave infrared polarization information image;
taking the infrared polarization information image of the current wave band as a reference, solving the deflection angle and the two-dimensional scaling of the infrared polarization information image of the other wave band, and correcting the long-wave infrared polarization information image of the other wave band according to the deflection angle and the two-dimensional scaling;
and performing autocorrelation calculation based on the extracted edge linear features, determining the two-dimensional offset of the infrared polarization information image of the current waveband according to the peak occurrence position, and performing translation adjustment on the infrared polarization information image of the other waveband according to the two-dimensional offset to realize registration of the infrared polarization information images of different wavebands.
In a third aspect of the present invention, an infrared polarization information fusion device based on wavelet packet transformation is provided, including: at least one processor, at least one memory and a computer program stored in the memory, which when executed by the processor, implement the method as previously described.
In a fourth aspect of the present invention, an infrared polarization information fusion system based on wavelet packet transformation is provided, which includes the infrared polarization information fusion device based on wavelet packet transformation as described above and at least two infrared polarization imaging measurement devices connected thereto.
In a fifth aspect of the invention, a computer-readable storage medium is provided, on which a computer program is stored, which, when being executed by a processor, carries out the method as set forth above.
The implementation of the infrared polarization information fusion method, the device, the equipment and the system based on the wavelet packet transformation and the computer readable storage medium has the following beneficial effects: according to the invention, aiming at the characteristic difference of the infrared polarization information of the target surface in the medium and long wave bands, the medium and long wave band infrared polarization degree images are fused based on a wavelet packet transformation method, and the contour characteristics such as high-frequency details and the like of the fused images are prominent, so that the accuracy of target detection and identification can be improved.
Drawings
FIG. 1a and FIG. 1b are the infrared polarization information images of the infrared standard plate in the medium wave band and the long wave band, respectively;
FIG. 2 is a flow chart of a method for infrared polarization information fusion based on wavelet packet transformation according to a preferred embodiment of the present invention;
FIGS. 3a and 3b are fused images obtained by the wavelet packet transform-based infrared polarization information fusion method according to the present invention;
fig. 4 is a block diagram of an infrared polarization information fusion device based on wavelet packet transformation according to a preferred embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention firstly analyzes the difference of the infrared polarization information of the target material as follows:
for an opaque radiation target, the radiation at its surface consists mainly of background reflections and self-radiation. According to the infrared polarization imaging radiation transmission model, the Stokes vector S of an opaque radiation object before the object is incident on a polarization deviceinExpressed as:
Sin＝Sr+Se＝∫f(θi,φi,θr,φr,λ)cos(θr)dΩr·Si+surf·Iobj(1)
in the formula SrAs target reflection Stokes vector, SeRadiating the Stokes vector, θ, for the target itselfiAt the angle of incidence zenith, θrTo reflect the zenith angle phiiIs the incident azimuth angle phirFor the reflection azimuth, λ is the detection wavelength, f (θ)i,φi,θr,φrλ) is the Bidirectional Reflectance Distribution Function (BRDF),surfis the surface emissivity of the material, Iobjd.OMEGA.for target radiation intensityiIs the angle of incidence. Expressing the transmission relation from incident radiation to reflected radiation by using a Fresnel reflection Muller matrix and a BRDF model based on micro-surface element theory, and according to the Stokes vector S of an incident interfaceiThe Stokes vector S of the reflection can be obtainedrCan be expressed as:
wherein
Respectively, represent the reflected Stokes parameters,
respectively representing the incident Stokes parameter, parameter S0Is related to the incident light intensity; s1Linear polarization information associated with the 0 ° and 90 ° directions; s2Linear polarization information associated with 45 ° and 135 ° directions; s3Associated with left-handed and right-handed circular polarization information, f00、f01、f02、f03、f10、f11、f12、f13、f20、 f21、f22、f23、f30、f31、f32、f33Is the corresponding value of the polarized BRDF model obtained by the action of the scalar BRDF function and the Muller matrix M of 4 × 4.
For an opaque radiation target surface, typically the radiation received by infrared polarized imaging includes radiation from the scene and reflections from other radiation sources, the total radiance at the target surface can be expressed as:
Ltol＝Le+Lr＝LB·surf+∫f(θi,φi,θr,φr,λ)·L(θi,φiλ)cos(θi)dΩi(3)
wherein L iseIs the target's own radiance, LrIs the brightness of the reflected radiation of the target surface, LBIs the black body radiation brightness corresponding to the spectral radiation brightness of the target, L (theta)i,φiAnd λ) is the incident radiation brightness.
Since the incident radiation is thermal radiation at the same temperature as the surface, the total radiation is black body radiation at the same temperature, i.e.
LB·0＝LB·surf+∫f(θi,φi,θr,φr,λ)×LB·0cos(θi)dΩi(4)
Wherein0Is a vector of the radiation rate of the black body,0＝(1,0,0,0)T。
the expression of the surface polarized radiation emissivity of the target material obtained by the formula (4) is as follows:
therefore, the Stokes vector S of the self radiation of the target material surface can be knowneCan be expressed as: se＝surf·IobjIn which IobjIs the radiation intensity of the target. Then equation (1) can be expressed as:
wherein S0、S1、S2And S3Representing Stokes parameters S0Image, S1Image, S2Image sum S3And (4) an image. I isbgIs the background radiation intensity.
The calculation expression of polarization degree based on the infrared radiation polarization transmission model can be derived from the definition of the formula (6) and the polarization degree as follows:
from the equation (7), the infrared polarization degree characteristic of the target material surface is closely related to the detection wavelength. Mainly, compared with the long wave band, the influence of the radiation of the surrounding environment in the medium wave band is more obvious, so the linear polarization degree of the target in the medium wave band is lower than that in the long wave band.
Referring to fig. 1a and 1b, the infrared polarization information images of the infrared standard plate in the medium wave band and the long wave band are shown, respectively. As can be seen from the figure: the difference between the middle wave polarization information image and the long wave infrared polarization information image is obvious when the edge of the infrared standard plate and the edge outline of the test support platform and other detailed information are different. Wherein the long wave infrared polarization characteristic of the edge profile of the infrared standard plate of FIG. 1b is more distinct than the medium wave infrared polarization characteristic of FIG. 1a, and the medium wave infrared polarization characteristic of the edge profile of the test support table of FIG. 1a is more distinct than the long wave infrared polarization characteristic of FIG. 1 b. In consideration of the difference of detail information such as target edge contour and the like in medium and long wave bands, in order to enhance the detectable identification characteristic of a target scene, the polarization characteristics of the medium and long waves need to be fused to present infrared polarization characteristic information of a target in the same image, and a new method is provided for target detection and identification.
Therefore, the invention provides an infrared polarization information fusion method based on wavelet packet transformation, so as to improve the accuracy of target identification and detection based on the infrared polarization profile characteristics of the target surface.
Fig. 2 is a flowchart of an infrared polarization information fusion method based on wavelet packet transformation according to a preferred embodiment of the present invention. As shown in fig. 2, the method for fusing infrared polarization information based on wavelet packet transformation provided in this embodiment includes the following steps:
first, in step S1, an image registration step is performed to acquire infrared polarization information images of different wavelength bands and perform registration. The infrared polarization information images of different wave bands comprise long-wave infrared polarization information images and medium-wave infrared polarization information images. The long wave means a wavelength of 8-12 μm, and the medium wave means a wavelength of 3-5 μm.
In the infrared polarization imaging test process, the mid-and long-wave infrared polarization information can generate offset on image representation due to the fact that the moving polarization and the like are generated in the using equipment difference or the mid-and long-wave infrared polarization imaging test process, and therefore the mid-and long-wave infrared polarization information image and the long-wave infrared polarization information image need to be registered. In view of the characteristic that the infrared polarization information image can highlight the edge contour features of the target, the infrared polarization information image is registered by adopting an image feature anchoring method.
Preferably, the step S1 includes the following specific steps:
(1) performing edge extraction on the infrared polarization information image, namely respectively extracting edge linear characteristics of targets in the long-wave infrared polarization information image and the medium-wave infrared polarization information image;
(2) solving the infrared polarization characteristics in the infrared polarization information images of different wave bands, taking the infrared polarization information image of the current wave band as a reference, solving the deflection angle and the two-dimensional scaling of the infrared polarization information image of the other wave band, and correcting the long-wave infrared polarization information image of the other wave band according to the deflection angle and the two-dimensional scaling; for example, the long-wave infrared polarization information image is corrected with reference to the medium-wave infrared polarization information image, or the medium-wave infrared polarization information image is corrected with reference to the long-wave infrared polarization information image.
(3) And performing self-correlation calculation based on edge linear features in infrared polarization information images of different wave bands, determining the two-dimensional offset of the infrared polarization information image of another wave band according to the peak occurrence position of the self-correlation calculation by taking the medium wave infrared polarization information image of the current wave band as a reference, and performing translation adjustment on the infrared polarization information image of another wave band according to the two-dimensional offset. For example, the long-wave infrared polarization information image is subjected to translational modulation with reference to the medium-wave infrared polarization information image, or the medium-wave infrared polarization information image is subjected to translational modulation with reference to the long-wave infrared polarization information image.
Subsequently, in step S2, an image fusion step is performed to perform wavelet packet decomposition on the registered infrared polarization information images of different wavebands, and to fuse corresponding band sub-images obtained by the decomposition of the medium-wave infrared polarization information image and the long-wave infrared polarization information image, and then to obtain a fused image through wavelet packet reconstruction. In the invention, the proper fusion rule is adopted to respectively perform targeted fusion processing on the sub-images of the frequency bands with different characteristics.
Low-frequency coefficient fusion rule: as can be seen from fig. 1a and 1b, after wavelet packet decomposition, the low frequency components of the image include most of the information of the original image, and have a great influence on the quality of fusion. Therefore, the low-frequency component fusion is preferably performed in a weighted average manner in this step:
WPl,F＝aWPl,A+bWPl,B(8)
wherein, WPi,FFor fusing the low frequency component part of the image, a and b are weighting coefficients, and a + b is 1, WPi,AFor low-frequency subimages, WP, obtained after decomposition of the mid-wave infrared polarization information imagei,BAs long-wave infrared polarized information imagesAnd (5) obtaining a low-frequency sub-image after decomposition. Since the source images to be fused are images from the same scene, the low frequency components are usually relatively close, and a is usually selected to be b. Subscripts A and B in the invention respectively represent infrared polarization information images to be fused, namely a medium wave infrared polarization information image and a long wave infrared polarization information image.
High-frequency coefficient fusion rule: for the high-frequency component, a fusion rule based on local region features is adopted. This is because the details of the image show that there is a strong correlation between adjacent pixels, so the values of other pixels in the adjacent regions need to be considered during the fusion. The local area feature is expressed by a weighted variance in the local area Q centered at (m, n), and the size of the local area G is selected to be 3 × 3, 5 × 5, 7 × 7, or the like, as the case may be. The feature quantity, i.e., the local area variance, can be expressed as
In the formula G1k(m, n) represents the local area variance with the point (m, n) as the center in the kth node coefficient matrix of the layer 1 decomposed by the infrared polarization information image to be fused; WP1kA k node coefficient matrix term representing a level 1 of the decomposition; w (q) represents a weight coefficient matrix, typically a gaussian weight coefficient matrix;
the k-th node coefficient matrix of the layer 1 representing the decomposition is the average of the local area Q centered at the point (m, n).
Therefore, the performing high-frequency component fusion by the fusion method based on local region features in the above steps specifically includes:
(1) calculating the variance matching degree M of the long wave infrared polarization information image and the medium wave infrared polarization information image and the local area corresponding to the kth node coefficient matrix of the 1 st layer1k,AB(m,n)：
Wherein, WP1k,A(m, n) is the k node coefficient matrix term of the 1 st layer decomposed by the medium wave infrared polarization information image, WP1k,B(m, n) is the k node coefficient matrix item of the 1 st layer decomposed for the long wave infrared polarization information image,
the average value of a Q area with a point (m, n) as the center is used as a k node coefficient matrix of a layer 1 decomposed by the medium wave infrared polarization information image;
the average value of a local area Q taking a point (m, n) as the center of a kth node coefficient matrix of a layer 1 decomposed on a long-wave infrared polarization information image; g1k,A(m, n) is the local region variance with the point (m, n) as the center in the kth node coefficient matrix of the layer 1 of the medium-wave infrared polarization information image decomposition; g1k,BAnd (m, n) is the local area variance with the point (m, n) as the center in the kth node coefficient matrix of the layer 1 decomposed by the long-wave infrared polarization information image. M1k,ABThe value of (m, n) varies between 0 and 1, and the smaller the value, the lower the correlation degree of the high-frequency coefficient matrixes of the two images.
(2) Judging variance matching degree M1k,AB(M, n) is less than a preset threshold T, if M is1k,ABIf < T, the k-th node coefficient matrix item WP of the layer 1 of the high-frequency component part of the fusion image1k,F(m, n) is:
if M is1k,ABNot less than T, then the k node coefficient matrix item WP of the layer 1 of the high-frequency component part of the fused image1k,F(m, n) is:
wherein, the weighting coefficient of the kth node of the 1 st layer
The preset threshold value T is a parameter calibrated according to experiments, and is generally 0.4-1. Attention is paid to: in order to ensure consistency of the fused high-frequency detail components, the high-frequency detail coefficient matrix can be adjusted according to 8-domain pixels of the fused image pixels.
After the high-frequency component and the low-frequency component are fused, a fused image can be obtained through wavelet packet reconstruction.
Finally, in step S3, a target recognition step is performed to perform target recognition based on the fused image. In this step, various methods familiar to and applicable by those skilled in the art can be adopted to perform target recognition on the fused image, and will not be described herein again.
The effect of the infrared polarization information fusion method based on wavelet packet transformation is verified and analyzed. The medium wave polarization information image and the long wave infrared polarization information image in fig. 1a and 1b are fused as original images. Fig. 3a and 3b are fused images obtained by the infrared polarization information fusion method based on wavelet packet transformation according to the present invention. Wherein, fig. 3a is the fusion by taking the long wave infrared polarization information image as a reference image, and fig. 3b is the fusion by taking the medium wave infrared polarization information image as a reference image. As can be seen from the figure, the fused image obtained by the infrared polarization information fusion method based on wavelet packet transformation has rich edge details. Therefore, the infrared polarization information fusion method based on wavelet packet transformation has good effect, the image is overall clearer, and details such as edge outline, radiation intensity information and the like of the target are clearer, thereby being beneficial to scene perception and target identification.
Based on the same inventive concept, the embodiment of the invention also provides an infrared polarization information fusion device based on wavelet packet transformation. Fig. 4 is a block diagram of an infrared polarization information fusion apparatus based on wavelet packet transformation according to a preferred embodiment of the present invention. As shown in fig. 4, this embodiment provides an infrared polarization information fusion apparatus 400 including: an image registration unit 401, an image fusion unit 402 and an object recognition unit 403.
And an image registration unit 401, configured to acquire infrared polarization information images of different wavebands and perform registration. The infrared polarization information images of different wave bands comprise long wave infrared polarization information images and medium wave infrared polarization information images.
Preferably, the image registration unit 401 performs registration by:
(1) performing edge extraction on the infrared polarization information image, namely respectively extracting edge linear characteristics of targets in the long-wave infrared polarization information image and the medium-wave infrared polarization information image;
(2) solving the infrared polarization characteristics in the infrared polarization information images of different wave bands, taking the infrared polarization information image of the current wave band as a reference, solving the deflection angle and the two-dimensional scaling of the infrared polarization information image of the other wave band, and correcting the long-wave infrared polarization information image of the other wave band according to the deflection angle and the two-dimensional scaling; for example, the long-wave infrared polarization information image is corrected with reference to the medium-wave infrared polarization information image, or the medium-wave infrared polarization information image is corrected with reference to the long-wave infrared polarization information image.
(3) And performing self-correlation calculation based on edge linear features in infrared polarization information images of different wave bands, determining the two-dimensional offset of the infrared polarization information image of another wave band according to the peak occurrence position of the self-correlation calculation by taking the medium wave infrared polarization information image of the current wave band as a reference, and performing translation adjustment on the infrared polarization information image of another wave band according to the two-dimensional offset. For example, the long-wave infrared polarization information image is subjected to translational modulation with reference to the medium-wave infrared polarization information image, or the medium-wave infrared polarization information image is subjected to translational modulation with reference to the long-wave infrared polarization information image.
And the image fusion unit 402 is configured to perform wavelet packet decomposition on the registered infrared polarization information images of different wavebands, fuse corresponding frequency band sub-images obtained by decomposing the medium-wave infrared polarization information image and the long-wave infrared polarization information image, and reconstruct the wavelet packet to obtain a fusion image. In the invention, the proper fusion rule is adopted to respectively perform targeted fusion processing on the sub-images of the frequency bands with different characteristics.
Preferably, the image fusion unit 402 performs low frequency component fusion by using the weighted average method, such as using equation (8). The image fusion unit 402 performs high-frequency component fusion by the above-described fusion method based on local region features.
And an object recognition unit 403, configured to perform object recognition based on the fused image.
The embodiment of the invention also provides equipment for executing the infrared polarization information fusion method based on the wavelet packet transformation, which comprises the following steps: at least one processor, at least one memory, and a computer program stored in the memory, which when executed by the processor, implement the method for wavelet packet transform based infrared polarization information fusion as in the above embodiments.
The embodiment of the invention provides an infrared polarization information fusion system based on wavelet packet transformation, which comprises the infrared polarization information fusion equipment based on the wavelet packet transformation and at least two infrared polarization imaging measurement devices connected with the infrared polarization information fusion equipment. These infrared polarization imaging measuring devices include an infrared polarization imaging measuring device for acquiring a long-wave infrared radiation image, and an infrared polarization imaging measuring device for acquiring a medium-wave infrared radiation image. Each infrared polarization imaging measurement device can acquire an infrared radiation image I of a target scene with the rotating polarization directions of 0 degree, 45 degrees, 90 degrees and 135 degrees through a rotating polaroid0°、I45°、I90°、I135°And respectively calculating to obtain a medium wave infrared polarization information image and a long wave infrared polarization information image.
An embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the infrared polarization information fusion method based on wavelet packet transformation as in the above embodiments is implemented.
The invention verifies the relevance between the detection wavelength and the infrared polarization characteristic through theoretical analysis, and the medium wave infrared polarization degree characteristic of the target surface is lower than the long wave infrared polarization degree characteristic because the surrounding environment has larger influence on the medium wave band range. Therefore, the method mainly aims at the difference rule of the infrared polarization characteristics of the target surface in the range of the medium-wave and long-wave bands, researches and analyzes the medium-wave and long-wave infrared polarization information fusion method, fuses the medium-wave infrared polarization information image and the long-wave infrared polarization information image through a wavelet packet transformation-based method, has prominent profile characteristics such as high-frequency details and the like of the fused images, and can provide new effective characteristics for target detection and identification.
It should be understood that the principle of the method and the device for fusing infrared polarization information based on wavelet packet transformation in the present invention is the same, so that the detailed description of the embodiment of the method for fusing infrared polarization information based on wavelet packet transformation is also applicable to the device for fusing infrared polarization information based on wavelet packet transformation.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may be modified or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions.
Claims (10)
1. An infrared polarization information fusion method based on wavelet packet transformation is characterized by comprising the following steps:
acquiring infrared polarization information images of different wave bands and registering; the infrared polarization information images of different wave bands comprise long wave infrared polarization information images and medium wave infrared polarization information images;
performing wavelet packet decomposition on the registered infrared polarization information images of different wave bands, fusing corresponding frequency band sub-images obtained by decomposing the medium wave infrared polarization information images and the long wave infrared polarization information images, and reconstructing through wavelet packets to obtain fused images;
and carrying out target identification based on the fused image.
2. The wavelet packet transformation-based infrared polarization information fusion method of claim 1, wherein the fusion of the corresponding band subimages decomposed from the medium-wave infrared polarization information image and the long-wave infrared polarization information image comprises low-frequency component fusion in a weighted average manner:
WPl,F＝aWPl,A+bWPl,B
wherein, WPi,FFor fusing the low frequency component part of the image, a and b are weighting coefficients, and a + b is 1, WPi,AFor low-frequency subimages, WP, obtained after decomposition of the mid-wave infrared polarization information imagei,BThe long-wave infrared polarization information image is decomposed to obtain a low-frequency sub-image.
3. The wavelet packet transformation-based infrared polarization information fusion method according to claim 1 or 2, wherein the fusing of the corresponding band subimages decomposed from the medium-wave infrared polarization information image and the long-wave infrared polarization information image comprises performing high-frequency component fusion by using a local region feature-based fusion method: calculating the variance matching degree M of the long wave infrared polarization information image and the medium wave infrared polarization information image and the local area corresponding to the kth node coefficient matrix of the 1 st layer1k,AB(m,n)：
Wherein, WP1k,A(m, n) is the k node coefficient matrix term of the 1 st layer decomposed by the medium wave infrared polarization information image, WP1k,B(m, n) is the k node coefficient matrix item of the 1 st layer decomposed for the long wave infrared polarization information image,
the average value of a Q area with a point (m, n) as the center is used as a k node coefficient matrix of a layer 1 decomposed by the medium wave infrared polarization information image;
the average value of a local area Q taking a point (m, n) as the center of a kth node coefficient matrix of a layer 1 decomposed on a long-wave infrared polarization information image; g1k,A(m, n) is the local area variance with the point (m, n) as the center in the kth node coefficient matrix of the layer 1 decomposed by the medium wave infrared polarization information image; g1k,B(m, n) is the local area variance with the point (m, n) as the center in the kth node coefficient matrix of the layer 1 decomposed by the long-wave infrared polarization information image;
judging variance matching degree M1k,ABIf (m, n) is less than the preset threshold, if yes, the kth node coefficient matrix term WP of the layer 1 of the high-frequency component part of the fused image is fused1k,F(m, n) is:
otherwise, fusing the kth node coefficient matrix term WP of the layer 1 of the high-frequency component part of the image1k,F(m, n) is:
wherein, the weighting coefficient of the kth node of the 1 st layer
4. The infrared polarization information fusion method based on the wavelet packet transform according to claim 1 or 2, wherein the acquiring and registering infrared polarization information images of different wave bands comprises:
performing edge extraction on the infrared polarization information image, and respectively extracting edge linear characteristics of targets in the long-wave infrared polarization information image and the medium-wave infrared polarization information image;
taking the infrared polarization information image of the current wave band as a reference, solving the deflection angle and the two-dimensional scaling of the infrared polarization information image of the other wave band, and correcting the long-wave infrared polarization information image of the other wave band according to the deflection angle and the two-dimensional scaling;
and performing autocorrelation calculation based on the extracted edge linear features, confirming the two-dimensional offset of the infrared polarization information image of the current waveband according to the peak occurrence position, and performing translation adjustment on the infrared polarization information image of the other waveband according to the two-dimensional offset to realize registration of the infrared polarization information images of different wavebands.
5. An infrared polarization information fusion device based on wavelet packet transformation is characterized by comprising:
the image registration unit is used for acquiring infrared polarization information images of different wave bands and registering the infrared polarization information images; the infrared polarization information images of different wave bands comprise long wave infrared polarization information images and medium wave infrared polarization information images;
the image fusion unit is used for performing wavelet packet decomposition on the registered infrared polarization information images of different wave bands, fusing corresponding frequency band sub-images obtained by decomposing the medium wave infrared polarization information images and the long wave infrared polarization information images, and reconstructing through wavelet packets to obtain a fusion image;
and the target identification unit is used for carrying out target identification based on the fused image.
6. The infrared polarization information fusion device based on wavelet packet transformation as claimed in claim 5, wherein said image fusion unit performs low frequency component fusion by means of weighted average:
WPl,F＝aWPl,A+bWPl,B
wherein, WPi,FFor fusing the low frequency component part of the image, a and b are weighting coefficients, and a + b is 1, WPi,AFor low-frequency subimages, WP, obtained after decomposition of the mid-wave infrared polarization information imagei,BThe long-wave infrared polarization information image is decomposed to obtain a low-frequency sub-image.
7. Wavelet packet transform-based according to claim 5 or 6The infrared polarization information fusion device is characterized in that the image fusion unit adopts a fusion method based on local region characteristics to perform high-frequency component fusion: calculating the variance matching degree M of the long wave infrared polarization information image and the medium wave infrared polarization information image and the local area corresponding to the kth node coefficient matrix of the 1 st layer1k,AB(m,n)：
Wherein, WP1k,A(m, n) is the k node coefficient matrix term of the 1 st layer decomposed by the medium wave infrared polarization information image, WP1k,B(m, n) is the k node coefficient matrix item of the 1 st layer decomposed for the long wave infrared polarization information image,
the average value of a Q area with a point (m, n) as the center is used as a k node coefficient matrix of a layer 1 decomposed by the medium wave infrared polarization information image;
the average value of a local area Q taking a point (m, n) as the center of a kth node coefficient matrix of a layer 1 decomposed on a long-wave infrared polarization information image; g1k,A(m, n) is the local area variance with the point (m, n) as the center in the kth node coefficient matrix of the layer 1 decomposed by the medium wave infrared polarization information image; g1k,B(m, n) is the local area variance with the point (m, n) as the center in the kth node coefficient matrix of the layer 1 decomposed by the long-wave infrared polarization information image;
judging variance matching degree M1k,ABIf (m, n) is less than the preset threshold, if yes, the kth node coefficient matrix term WP of the layer 1 of the high-frequency component part of the fused image is fused1k,F(m, n) is:
otherwise, fusing the kth node coefficient matrix term WP of the layer 1 of the high-frequency component part of the image1k,F(m, n) is:
wherein, the weighting coefficient of the kth node of the 1 st layer
8. An infrared polarization information fusion device based on wavelet packet transformation, comprising: at least one processor, at least one memory and a computer program stored in the memory, characterized in that the computer program realizes the method according to any one of claims 1 to 4 when the computer program is executed by the processor.
9. An infrared polarization information fusion system based on wavelet packet transformation, which is characterized by comprising the infrared polarization information fusion device based on wavelet packet transformation according to claim 8 and at least two infrared polarization imaging measurement devices connected with the infrared polarization information fusion device.
10. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the method according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010507681.6A CN111667517A (en) | 2020-06-05 | 2020-06-05 | Infrared polarization information fusion method and device based on wavelet packet transformation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010507681.6A CN111667517A (en) | 2020-06-05 | 2020-06-05 | Infrared polarization information fusion method and device based on wavelet packet transformation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111667517A true CN111667517A (en) | 2020-09-15 |
Family
ID=72386909
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010507681.6A Pending CN111667517A (en) | 2020-06-05 | 2020-06-05 | Infrared polarization information fusion method and device based on wavelet packet transformation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111667517A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113421206A (en) * | 2021-07-16 | 2021-09-21 | 合肥工业大学 | Image enhancement method based on 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 |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102567977A (en) * | 2011-12-31 | 2012-07-11 | 南京理工大学 | Self-adaptive fusing method of infrared polarization image based on wavelets |
| CN105427268A (en) * | 2015-12-01 | 2016-03-23 | 中国航空工业集团公司洛阳电光设备研究所 | Medium-long-wave dual-waveband infrared image feature level color fusion method |
| CN106846289A (en) * | 2017-01-17 | 2017-06-13 | 中北大学 | A kind of infrared light intensity and polarization image fusion method based on conspicuousness migration with details classification |
| CN107633495A (en) * | 2017-08-02 | 2018-01-26 | 中北大学 | A kind of infrared polarization based on complementary relationship and the more embedded fusion methods of algorithm 2D VMD of intensity image |
| CN108364277A (en) * | 2017-12-20 | 2018-08-03 | 南昌航空大学 | A kind of infrared small target detection method of two-hand infrared image fusion |
| CN108492274A (en) * | 2018-04-03 | 2018-09-04 | 中国人民解放军国防科技大学 | Long-wave infrared polarization feature extraction and fusion image enhancement method |
| CN109741376A (en) * | 2018-11-19 | 2019-05-10 | 天津津航技术物理研究所 | It is a kind of based on improve RANSAC algorithm in, LONG WAVE INFRARED method for registering images |
| CN110472658A (en) * | 2019-07-05 | 2019-11-19 | 哈尔滨工程大学 | A kind of the level fusion and extracting method of the detection of moving-target multi-source |
-
2020
- 2020-06-05 CN CN202010507681.6A patent/CN111667517A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102567977A (en) * | 2011-12-31 | 2012-07-11 | 南京理工大学 | Self-adaptive fusing method of infrared polarization image based on wavelets |
| CN105427268A (en) * | 2015-12-01 | 2016-03-23 | 中国航空工业集团公司洛阳电光设备研究所 | Medium-long-wave dual-waveband infrared image feature level color fusion method |
| CN106846289A (en) * | 2017-01-17 | 2017-06-13 | 中北大学 | A kind of infrared light intensity and polarization image fusion method based on conspicuousness migration with details classification |
| CN107633495A (en) * | 2017-08-02 | 2018-01-26 | 中北大学 | A kind of infrared polarization based on complementary relationship and the more embedded fusion methods of algorithm 2D VMD of intensity image |
| CN108364277A (en) * | 2017-12-20 | 2018-08-03 | 南昌航空大学 | A kind of infrared small target detection method of two-hand infrared image fusion |
| CN108492274A (en) * | 2018-04-03 | 2018-09-04 | 中国人民解放军国防科技大学 | Long-wave infrared polarization feature extraction and fusion image enhancement method |
| CN109741376A (en) * | 2018-11-19 | 2019-05-10 | 天津津航技术物理研究所 | It is a kind of based on improve RANSAC algorithm in, LONG WAVE INFRARED method for registering images |
| CN110472658A (en) * | 2019-07-05 | 2019-11-19 | 哈尔滨工程大学 | A kind of the level fusion and extracting method of the detection of moving-target multi-source |
Non-Patent Citations (7)
| Title |
|---|
| 姚敏;: "基于特征点自动匹配的红外图像配准研究" * |
| 孙玉秋,田岩,彭代强,田金文: "双波段红外图像融合的小波分维算法" * |
| 孙玉秋;田金文;柳健;: "基于小波变换的双色红外图像融合检测方法" * |
| 张晶晶;方勇华;: "基于可见光的多波段偏振图像融合新算法" * |
| 张雨晨;李江勇;: "基于小波变换的中波红外偏振图像融合" * |
| 杨风暴;倪国强;张雷;: "红外中波细分图像的小波包变换融合研究" * |
| 陈伟力;王霞;金伟其;李福文;: "基于小波包变换的中波红外偏振图像融合研究" * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113421206A (en) * | 2021-07-16 | 2021-09-21 | 合肥工业大学 | Image enhancement method based on 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 |
| CN113421205B (en) * | 2021-07-16 | 2022-11-15 | 合肥工业大学 | Small target detection method combined with infrared polarization imaging |
| CN113421206B (en) * | 2021-07-16 | 2022-11-15 | 合肥工业大学 | Image enhancement method based on infrared polarization imaging |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Li et al. | Haze and thin cloud removal via sphere model improved dark channel prior | |
| CN111667516A (en) | Infrared polarization information fusion method based on Laplacian pyramid decomposition structure | |
| CN108492274B (en) | Long-wave infrared polarization feature extraction and fusion image enhancement method | |
| US7548661B2 (en) | Single-image vignetting correction | |
| US11068700B2 (en) | Polarization imaging for facial recognition enhancement system and method | |
| US6738532B1 (en) | Image registration using reduced resolution transform space | |
| EP2561482B1 (en) | Shape and photometric invariants recovery from polarisation images | |
| CN100573193C (en) | High light spectrum small target detection method and device | |
| CN111667517A (en) | Infrared polarization information fusion method and device based on wavelet packet transformation | |
| KR101928391B1 (en) | Method and apparatus for data fusion of multi spectral image and radar image | |
| CN110520768B (en) | Hyperspectral light field imaging method and system | |
| Yusuf et al. | Spectral information analysis of image fusion data for remote sensing applications | |
| CN112215876B (en) | Double-spectrum image registration fusion method, device, equipment and storage medium | |
| Chen et al. | Adaptive wavelet filter with edge compensation for remote sensing image denoising | |
| CN103839077A (en) | Low-rank-represented polarization SAR image classification method based on superpixel features | |
| CN111667519A (en) | Registration method and device for polarized images with different fields of view | |
| Shah et al. | Multimodal image/video fusion rule using generalized pixel significance based on statistical properties of the neighborhood | |
| Xu et al. | A correlation-based interpolation algorithm for division-of-focal-plane polarization sensors | |
| CN109801253B (en) | Self-adaptive cloud area detection method for high-resolution optical remote sensing image | |
| KR101928393B1 (en) | Method and apparatus for data fusion of infrared image and radar image | |
| CN113506330A (en) | Different-waveband polarization angle image fusion method based on multi-scale transformation | |
| Temel et al. | BLeSS: Bio-inspired low-level spatiochromatic similarity assisted image quality assessment | |
| Liu et al. | Multitask saliency detection model for synthetic aperture radar (SAR) image and its application in SAR and optical image fusion | |
| Zhao | Research on multiband packet fusion algorithm for hyperspectral remote sensing images | |
| Abramova et al. | Improvement of terahertz images by adaptive discrete cosine transform (DCT)-based denoising |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200915 |