CN107451984B - Infrared and visible light image fusion algorithm based on mixed multi-scale analysis - Google Patents
Infrared and visible light image fusion algorithm based on mixed multi-scale analysis Download PDFInfo
- Publication number
- CN107451984B CN107451984B CN201710621620.0A CN201710621620A CN107451984B CN 107451984 B CN107451984 B CN 107451984B CN 201710621620 A CN201710621620 A CN 201710621620A CN 107451984 B CN107451984 B CN 107451984B
- Authority
- CN
- China
- Prior art keywords
- image
- band
- frequency sub
- frequency
- sub
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 230000004927 fusion Effects 0.000 title claims abstract description 37
- 238000000034 method Methods 0.000 claims abstract description 19
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 14
- 230000009466 transformation Effects 0.000 claims abstract description 11
- 230000003068 static effect Effects 0.000 claims abstract description 7
- 238000005259 measurement Methods 0.000 claims description 19
- 210000002569 neuron Anatomy 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 5
- 238000005070 sampling Methods 0.000 claims description 4
- 230000008602 contraction Effects 0.000 claims description 3
- 238000010304 firing Methods 0.000 claims description 3
- 230000001360 synchronised effect Effects 0.000 claims description 3
- 230000000007 visual effect Effects 0.000 claims description 3
- 230000006870 function Effects 0.000 description 8
- 230000000694 effects Effects 0.000 description 4
- 238000011156 evaluation Methods 0.000 description 4
- 230000007547 defect Effects 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 244000096108 cunha Species 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000007500 overflow downdraw method Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
- 230000016776 visual perception Effects 0.000 description 1
Images
Classifications
-
- 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
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10052—Images from lightfield camera
-
- 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]
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
The invention discloses an infrared and visible light image fusion algorithm based on mixed multi-scale analysis, which comprises the following steps: step 1: performing NSCT decomposition on the infrared and visible light images to obtain a low-frequency sub-band and a high-frequency sub-band; step 2: adopting static wavelet transformation to the low-frequency sub-band to obtain a low-frequency sub-band and three high-frequency sub-bands, and fusing the low-frequency sub-band and the high-frequency sub-bands respectively by adopting local energy and absolute value combination and a compressive sensing theory; and step 3: judging the definition of the image to be fused, and selecting the number of the enhancement layers of the LSCN according to a judgment criterion; and 4, step 4: adopting a fusion rule that the absolute value of the highest-layer high-frequency sub-band is larger, and adopting an improved PCNN model to fuse the other sub-bands; and 5: and performing NSCT inverse transformation on the fusion result to obtain a final fusion image. The fused image obtained by the method has the advantages of prominent edge, high contrast, prominent target and higher indexes such as average gradient, spatial frequency and the like of the algorithm than the prior art.
Description
Technical Field
The invention belongs to the technical field of image processing, and particularly relates to an infrared and visible light image fusion algorithm based on mixed multi-scale analysis.
Background
The image fusion method based on wavelet transform is a classic fusion algorithm, but wavelets can only represent isotropic objects, and are not an ideal representation tool for characteristics such as image central lines, edges and the like. Contourlet has wide application in image fusion, and through multi-scale and multi-direction decomposition of images, Contourlet can well capture detail characteristics in the images, thereby making up the defects of wavelets. However, since the Contourlet transform uses a down-sampling operation, it has no translation invariance, and is prone to generate a pseudo Gibbs phenomenon in image processing.
The nonsampled Contourlet transform (NSCT) proposed by a.l.cunha et al has translational invariance, can fully retain effective information of an image and generate a better fusion effect, but has the problems of poor sparsity of low-frequency partial images, unfavorable feature extraction and the like.
Disclosure of Invention
Aiming at the defects of the prior art, the invention solves the problems of low contrast, insufficient retention of edge information and the like in the fusion of infrared and visible light images.
In order to solve the technical problems, the technical scheme adopted by the invention is an infrared and visible light image fusion algorithm based on mixed multi-scale analysis, which comprises the following steps:
step 1: respectively carrying out NSCT decomposition on the infrared image and the visible light image to obtain a low-frequency sub-band LJ(x, y) and the high frequency subband Hj,r(x, y), wherein J is the number of decomposition layers, and J and r represent the decomposition scale and the direction number.
Step 2: and performing static wavelet transformation on the low-frequency sub-band to obtain a low-frequency sub-band and three high-frequency sub-bands, fusing the low-frequency sub-band and the high-frequency sub-band by respectively adopting local energy and absolute value combination and a compressive sensing theory, and performing wavelet inverse transformation to obtain the NSCT reconstructed low-frequency sub-band.
The method for fusing the low-frequency sub-bands by combining local energy and absolute value maximization and adopting a compressed sensing theory comprises the following specific steps of:
where EN is the local area energy, which is defined as:
the method adopts the combination of local energy and absolute value taking and the compressed sensing theory to fuse the high-frequency sub-band, and comprises the following specific steps:
1) image of high frequency sub-band with size of m × nAndare decomposed into sub-blocks of the same size, where j is not overlappedThinning each block of sub-block image using sym8 wavelet basis 1,2, 3;
2) designing a measurement matrix phi, and measuring and sampling the input high-frequency sub-band coefficient by using the measurement matrix to obtain a measurement vectorAndwherein k is 1, 2.., mxn;
3) calculating a measurement vectorAndstandard Deviation of (SD)kAnd definition EAVkAnd obtaining a fused measurement vector by adopting a fusion rule based on the combination of the regional standard deviation, the regional definition and the S function, namely:
the image standard deviation formula is:
The image sharpness formula is:
the weighting coefficient ω is obtained by an S-function, which is:
wherein,
f is a contraction factor of the S function, is greater than or equal to 1, and is taken as 5;
4) for the fused measurement vectorSparse reconstruction is carried out, and the reconstruction algorithm adopts OMP (object-to-process) so as to obtain the high-frequency sub-band of the fused image
SW to be obtainedFAndand performing static wavelet reconstruction to obtain a low-frequency sub-band finally used for NSCT reconstruction.
And step 3: judging the definition of the image to be fused, and selecting the number of the enhancement layers of the LSCN according to a judgment criterion, wherein the specific method comprises the following steps:
the image sharpness formula is:
calculating the image definition according to the formula (8), comparing the image definition with a threshold lambda, and determining the number of layers of high-frequency coefficient enhancement according to the comparison result, namely:
wherein J is the number of decomposition layers, S is the comprehensive definition of the source image, and alpha is taken1=α2=0.5,λ=27。
And 4, step 4: and (3) adopting a fusion rule that the absolute value of the highest-layer high-frequency sub-band is larger, and adopting an improved PCNN model to fuse the other sub-bands, wherein the specific fusion rule is as follows:
in order to improve the visual impression of the image, the other subbands except the highest-layer high-frequency subband n are fused by adopting an improved PCNN model, and a fusion coefficient is determined by comparing the sum of the firing amplitudes of the PCNN neurons, namely:
wherein M isijAnd (n) is the sum of the pulse ignition amplitudes output by the PCNN, j is 1,2, and n-1, wherein epsilon is a custom threshold, and epsilon is 0.002.
Because the output of the traditional PCNN adopts a hard limiting function and cannot reflect the amplitude difference of neuron ignition, the invention adopts a Sigmoid function as the output of the PCNN, which can better depict the difference of the ignition amplitude when the synchronous pulse is excited, and the output of the PCNN is defined as follows:
in order to better represent the edge information of the image, modified laplacian energy (SML) and local spatial frequency are selected as external input and linking coefficients of the PCNN, respectively. SML is defined as follows:
the spatial frequency is:
wherein RF, CF, MDF and SDF represent row frequency, column frequency, main diagonal frequency and sub diagonal frequency, respectively, and the formula is as follows:
and 5: and performing NSCT inverse transformation on the low-frequency sub-band and the high-frequency sub-band obtained by fusion to obtain a final fusion image.
The fused image obtained by the technical scheme of the invention has the advantages of prominent edge, higher contrast and brightness, prominent target, and higher average gradient, spatial frequency, standard deviation and information entropy of the algorithm than the method in the prior art, thereby effectively retaining the infrared target, effectively acquiring the spatial domain information of the source image and obtaining better fusion effect.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a source infrared image according to an embodiment;
FIG. 3 is a source visible image according to one embodiment;
fig. 4 is a fused image obtained in example one document 6;
fig. 5 is a fused image obtained in example one document 8;
fig. 6 is a fused image obtained in example one document 12;
FIG. 7 is a fused image obtained by an algorithm of the present invention according to an embodiment of the present invention;
FIG. 8 is an example two-source infrared image;
FIG. 9 is a visible light image from a second source according to an embodiment;
fig. 10 is a fused image obtained in example two document 6;
fig. 11 is a fused image obtained in example two document 8;
fig. 12 is a fused image obtained in example two document 12;
FIG. 13 is a fused image obtained by the algorithm of the second embodiment of the present invention.
Detailed Description
The following description will be made with reference to the accompanying drawings and examples, but the present invention is not limited thereto.
FIG. 1 shows the process of the present invention, an infrared and visible light image fusion algorithm based on mixed multi-scale analysis, comprising the following steps:
step 1: respectively carrying out NSCT decomposition on the infrared image and the visible light image to obtain a low-frequency sub-band LJ(x, y) and the high frequency subband Hj,r(x, y), wherein J is the number of decomposition layers, and J and r represent the decomposition scale and the direction number.
Step 2: and performing static wavelet transformation on the low-frequency sub-band to obtain a low-frequency sub-band and three high-frequency sub-bands, fusing the low-frequency sub-band and the high-frequency sub-band by respectively adopting local energy and absolute value combination and a compressive sensing theory, and performing wavelet inverse transformation to obtain the NSCT reconstructed low-frequency sub-band.
The method for fusing the low-frequency sub-bands by combining local energy and absolute value maximization and adopting a compressed sensing theory comprises the following specific steps of:
where EN is the local area energy, which is defined as:
the method adopts the combination of local energy and absolute value taking and the compressed sensing theory to fuse the high-frequency sub-band, and comprises the following specific steps:
1) image of high frequency sub-band with size of m × nAnddecomposing the images into sub-blocks which are not overlapped with each other and have the same size, wherein j is 1,2 and 3, and thinning each sub-block image by using sym8 wavelet base;
2) designing a measurement matrix phi by applying measurement momentsThe array carries out measurement sampling on the input high-frequency sub-band coefficient to obtain a measurement vectorAndwherein k is 1, 2.., mxn;
3) calculating a measurement vectorAndstandard Deviation of (SD)kAnd definition EAVkAnd obtaining a fused measurement vector by adopting a fusion rule based on the combination of the regional standard deviation, the regional definition and the S function, namely:
the image standard deviation formula is:
The image sharpness formula is:
the weighting coefficient ω is obtained by an S-function, which is:
wherein,
f is a contraction factor of the S function, is greater than or equal to 1, and is taken as 5;
4) for the fused measurement vectorSparse reconstruction is carried out, and the reconstruction algorithm adopts OMP (object-to-process) so as to obtain the high-frequency sub-band of the fused image
SW to be obtainedFAndand performing static wavelet reconstruction to obtain a low-frequency sub-band finally used for NSCT reconstruction.
And step 3: judging the definition of the image to be fused, and selecting the number of the enhancement layers of the LSCN according to a judgment criterion, wherein the specific method comprises the following steps:
the image sharpness formula is:
calculating the image definition according to the formula (8), comparing the image definition with a threshold lambda, and determining the number of layers of high-frequency coefficient enhancement according to the comparison result, namely:
wherein J is the number of decomposition layers, S is the comprehensive definition of the source image, and alpha is taken1=α2=0.5,λ=27。
And 4, step 4: and (3) adopting a fusion rule that the absolute value of the highest-layer high-frequency sub-band is larger, and adopting an improved PCNN model to fuse the other sub-bands, wherein the specific fusion rule is as follows:
in order to improve the visual impression of the image, the other subbands except the highest-layer high-frequency subband n are fused by adopting an improved PCNN model, and a fusion coefficient is determined by comparing the sum of the firing amplitudes of the PCNN neurons, namely:
wherein M isijAnd (n) is the sum of the pulse ignition amplitudes output by the PCNN, j is 1,2, and n-1, wherein epsilon is a custom threshold, and epsilon is 0.002.
Because the output of the traditional PCNN adopts a hard limiting function and cannot reflect the amplitude difference of neuron ignition, the invention adopts a Sigmoid function as the output of the PCNN, which can better depict the difference of the ignition amplitude when the synchronous pulse is excited, and the output of the PCNN is defined as follows:
in order to better represent the edge information of the image, modified laplacian energy (SML) and local spatial frequency are selected as external input and linking coefficients of the PCNN, respectively. SML is defined as follows:
the spatial frequency is:
wherein RF, CF, MDF and SDF represent row frequency, column frequency, main diagonal frequency and sub diagonal frequency, respectively, and the formula is as follows:
and 5: and performing NSCT inverse transformation on the low-frequency sub-band and the high-frequency sub-band obtained by fusion to obtain a final fusion image.
The infrared image records the infrared radiation information of a target object, has strong identification capability on a low-illumination or disguised target, but is not sensitive to brightness change. The visible light image is influenced to a greater extent by illumination, and can provide detailed information of a target scene. Therefore, the infrared image and the visible light image are fused, and a complementary image with clear background and prominent target can be obtained by combining the advantages of the infrared image and the visible light image, so that an observer can more accurately and comprehensively describe the scene.
The experimental data for the first and second examples are as follows:
fig. 2 is a source infrared image of the first embodiment, fig. 3 is a source visible light image of the first embodiment, the data of fig. 4 is document 6 of table 1, the data of fig. 5 is document 8 of table 1, the data of fig. 6 is document 12 of table 1, and the data of fig. 7 is the inventive algorithm of table 1.
Fig. 8 is a source infrared image of the second embodiment, fig. 9 is a source visible light image of the second embodiment, the data of fig. 10 is document 6 of table 2, the data of fig. 11 is document 8 of table 2, the data of fig. 12 is document 12 of table 2, and the data of fig. 13 is the inventive algorithm of table 2.
Objective evaluation analysis, it can be seen from tables 1 and 2 that each evaluation index of the method provided by the present embodiment is superior to other methods, and it is known from the above that the fusion effect of the present embodiment more conforms to human visual perception.
Table 1 first set of image fusion results evaluation:
table 2 second set of image fusion results evaluation:
the fused image obtained by the technical scheme of the invention has the advantages of prominent edge, higher contrast and brightness, prominent target, and higher average gradient, spatial frequency, standard deviation and information entropy of the algorithm than the method in the prior art, thereby effectively retaining the infrared target, effectively acquiring the spatial domain information of the source image and obtaining better fusion effect.
The embodiments of the present invention have been described in detail with reference to the accompanying drawings, but the present invention is not limited to the described embodiments. It will be apparent to those skilled in the art that various changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention.
Claims (1)
1. An infrared and visible light image fusion algorithm based on mixed multi-scale analysis is characterized in that: the method comprises the following steps:
step 1: respectively carrying out NSCT decomposition on the infrared image and the visible light image to obtain a low-frequency sub-band LJ(x, y) and the high frequency subband Hj,r(x, y), wherein J is the number of decomposition layers, and J and r represent the decomposition scale and the direction number;
step 2: adopting static wavelet transformation to the low-frequency sub-band to obtain a low-frequency sub-band and three high-frequency sub-bands, respectively adopting local energy and absolute value combination and compressive sensing theory to fuse the low-frequency sub-band and the high-frequency sub-band, and then carrying out wavelet inverse transformation to obtain a NSCT reconstructed low-frequency sub-band;
and step 3: judging the definition of the image to be fused, and selecting the number of the enhancement layers of the LSCN according to a judgment criterion;
and 4, step 4: adopting a fusion rule that the absolute value of the highest-layer high-frequency sub-band is larger, and adopting an improved PCNN model to fuse the other sub-bands;
and 5: performing NSCT inverse transformation on the low-frequency sub-band and the high-frequency sub-band obtained by fusion to obtain a final fusion image;
in step 2, the low-frequency sub-bands are fused by combining the local energy and the absolute value, and the specific method is as follows:
where EN is the local area energy, which is defined as:
in step 2, the high-frequency sub-bands are fused by adopting a compressed sensing theory, and the method specifically comprises the following steps:
1) image of high frequency sub-band with size of m × nAnddecomposing the images into sub-blocks which are not overlapped with each other and have the same size, wherein j is 1,2 and 3, and thinning each sub-block image by using sym8 wavelet base;
2) designing a measurement matrix phi, and measuring and sampling the input high-frequency sub-band coefficient by using the measurement matrix to obtain a measurement vectorAndwherein k is 1, 2.., mxn;
3) calculating a measurement vectorAndstandard Deviation of (SD)kAnd definition EAVkAnd obtaining a fused measurement vector by adopting a fusion rule based on the combination of the regional standard deviation, the regional definition and the S function, namely:
the image standard deviation formula is:
The image sharpness formula is:
the weighting coefficient ω is obtained by an S-function, which is:
wherein,
f is a contraction factor of the S function, is greater than or equal to 1, and is taken as 5;
4) for the fused measurement vectorSparse reconstruction is carried out, and the reconstruction algorithm adopts OMP (object-to-process) so as to obtain the high-frequency sub-band of the fused image
SW to be obtainedFAndperforming static wavelet reconstruction to obtain a low-frequency sub-band finally used for NSCT reconstruction;
in step 3, the specific method is as follows:
the image sharpness formula is:
calculating the image definition according to the formula (8), comparing the image definition with a threshold lambda, and determining the number of layers of high-frequency coefficient enhancement according to the comparison result, namely:
wherein J is the number of decomposition layers, S is the comprehensive definition of the source image, and alpha is taken1=α2=0.5,λ=27;
In step 4, the specific fusion rule is as follows:
in order to improve the visual impression of the image, the other subbands except the highest-layer high-frequency subband n are fused by adopting an improved PCNN model, and a fusion coefficient is determined by comparing the sum of the firing amplitudes of the PCNN neurons, namely:
wherein M isij(n) is the sum of pulse ignition amplitudes output by the PCNN, j is 1,2, and n-1, wherein epsilon is a self-defined threshold, and epsilon is 0.002;
because the output of the traditional PCNN adopts a hard limiting function, the amplitude difference of neuron ignition cannot be reflected, and the difference of the ignition amplitude when the synchronous pulse is excited can be better described by adopting a Sigmoid function as the output of the PCNN, wherein the output of the PCNN is defined as follows:
in order to better represent the edge information of the image, the improved Laplace energy SML and the local spatial frequency are respectively used as the external input and the link coefficient of the PCNN;
SML is defined as follows:
the spatial frequency is:
wherein RF, CF, MDF and SDF represent row frequency, column frequency, main diagonal frequency and sub diagonal frequency, respectively, and the formula is as follows:
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710621620.0A CN107451984B (en) | 2017-07-27 | 2017-07-27 | Infrared and visible light image fusion algorithm based on mixed multi-scale analysis |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710621620.0A CN107451984B (en) | 2017-07-27 | 2017-07-27 | Infrared and visible light image fusion algorithm based on mixed multi-scale analysis |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107451984A CN107451984A (en) | 2017-12-08 |
CN107451984B true CN107451984B (en) | 2021-06-22 |
Family
ID=60489702
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710621620.0A Active CN107451984B (en) | 2017-07-27 | 2017-07-27 | Infrared and visible light image fusion algorithm based on mixed multi-scale analysis |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107451984B (en) |
Families Citing this family (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108389158A (en) * | 2018-02-12 | 2018-08-10 | 河北大学 | A kind of infrared and visible light image interfusion method |
CN108648174A (en) * | 2018-04-04 | 2018-10-12 | 上海交通大学 | A kind of fusion method of multilayer images and system based on Autofocus Technology |
CN109410157B (en) * | 2018-06-19 | 2022-02-08 | 昆明理工大学 | Image fusion method based on low-rank sparse decomposition and PCNN |
CN109118460B (en) * | 2018-06-27 | 2020-08-11 | 河海大学 | Method and system for synchronously processing light-splitting polarization spectrum information |
CN109166088B (en) * | 2018-07-10 | 2022-01-28 | 南京理工大学 | Dual-waveband gray molten pool image fusion method based on non-downsampling wavelet transform |
CN109035189B (en) * | 2018-07-17 | 2021-07-23 | 桂林电子科技大学 | Infrared and weak visible light image fusion method based on Cauchy fuzzy function |
CN109191417A (en) * | 2018-09-11 | 2019-01-11 | 中国科学院长春光学精密机械与物理研究所 | It is detected based on conspicuousness and improves twin-channel method for self-adaption amalgamation and device |
CN109345788A (en) * | 2018-09-26 | 2019-02-15 | 国网安徽省电力有限公司铜陵市义安区供电公司 | A kind of monitoring early-warning system of view-based access control model feature |
CN109242815B (en) * | 2018-09-28 | 2022-03-18 | 合肥英睿***技术有限公司 | Infrared light image and visible light image fusion method and system |
CN109360182A (en) * | 2018-10-31 | 2019-02-19 | 广州供电局有限公司 | Image interfusion method, device, computer equipment and storage medium |
CN109978802B (en) * | 2019-02-13 | 2023-01-17 | 中山大学 | NSCT (non-subsampled Contourlet transform) and PCNN (pulse coupled neural network) -based high-dynamic-range image fusion method in compressed sensing domain |
CN110110786B (en) * | 2019-05-06 | 2023-04-14 | 电子科技大学 | Infrared and visible light image fusion method based on NSCT and DWT |
CN111861957B (en) * | 2020-07-02 | 2024-03-08 | Tcl华星光电技术有限公司 | Image fusion method and device |
CN112734683B (en) * | 2021-01-07 | 2024-02-20 | 西安电子科技大学 | Multi-scale SAR and infrared image fusion method based on target enhancement |
CN114359687B (en) * | 2021-12-07 | 2024-04-09 | 华南理工大学 | Target detection method, device, equipment and medium based on multi-mode data double fusion |
CN114757895B (en) * | 2022-03-25 | 2023-05-02 | 国网浙江省电力有限公司电力科学研究院 | Method and system for judging direct sunlight interference of infrared image of composite insulator |
CN116403057B (en) * | 2023-06-09 | 2023-08-18 | 山东瑞盈智能科技有限公司 | Power transmission line inspection method and system based on multi-source image fusion |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102254314A (en) * | 2011-07-17 | 2011-11-23 | 西安电子科技大学 | Visible-light/infrared image fusion method based on compressed sensing |
WO2016050290A1 (en) * | 2014-10-01 | 2016-04-07 | Metaio Gmbh | Method and system for determining at least one property related to at least part of a real environment |
CN105719263A (en) * | 2016-01-22 | 2016-06-29 | 昆明理工大学 | Visible light and infrared image fusion algorithm based on NSCT domain bottom layer visual features |
CN106600572A (en) * | 2016-12-12 | 2017-04-26 | 长春理工大学 | Adaptive low-illumination visible image and infrared image fusion method |
CN106981057A (en) * | 2017-03-24 | 2017-07-25 | 中国人民解放军国防科学技术大学 | A kind of NSST image interfusion methods based on RPCA |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1873693B (en) * | 2006-06-27 | 2010-05-12 | 上海大学 | Method based on Contourlet transformation, modified type pulse coupling neural network, and image amalgamation |
US9299130B2 (en) * | 2013-03-15 | 2016-03-29 | Trustees Of Tufts College | Methods and apparatus for image processing and analysis |
CN106327459B (en) * | 2016-09-06 | 2019-03-12 | 四川大学 | Visible light and infrared image fusion method based on UDCT and PCNN |
-
2017
- 2017-07-27 CN CN201710621620.0A patent/CN107451984B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102254314A (en) * | 2011-07-17 | 2011-11-23 | 西安电子科技大学 | Visible-light/infrared image fusion method based on compressed sensing |
WO2016050290A1 (en) * | 2014-10-01 | 2016-04-07 | Metaio Gmbh | Method and system for determining at least one property related to at least part of a real environment |
CN105719263A (en) * | 2016-01-22 | 2016-06-29 | 昆明理工大学 | Visible light and infrared image fusion algorithm based on NSCT domain bottom layer visual features |
CN106600572A (en) * | 2016-12-12 | 2017-04-26 | 长春理工大学 | Adaptive low-illumination visible image and infrared image fusion method |
CN106981057A (en) * | 2017-03-24 | 2017-07-25 | 中国人民解放军国防科学技术大学 | A kind of NSST image interfusion methods based on RPCA |
Non-Patent Citations (2)
Title |
---|
A fusion Algorithm for infrared and visible images based on RDU-PCNN and ICA-bases in NSST domain;Zhanwen Liu 等;《Infrared Physics & Technology》;20161031;第183-190页 * |
一种基于NSST和字典学习的红外和可见光图像融合算法;刘战文 等;《西北工业大学学报》;20170630;第35卷(第3期);第408-413页 * |
Also Published As
Publication number | Publication date |
---|---|
CN107451984A (en) | 2017-12-08 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107451984B (en) | Infrared and visible light image fusion algorithm based on mixed multi-scale analysis | |
Adu et al. | Image fusion based on nonsubsampled contourlet transform for infrared and visible light image | |
CN106846289B (en) | A kind of infrared light intensity and polarization image fusion method | |
CN104268847B (en) | A kind of infrared and visible light image fusion method based on interaction non-local mean filtering | |
CN109886908B (en) | Infrared image and visible light image fusion method | |
CN104123705B (en) | A kind of super-resolution rebuilding picture quality Contourlet territory evaluation methodology | |
CN107730482A (en) | A kind of sparse blending algorithm based on region energy and variance | |
CN105976346A (en) | Infrared and visible light image fusion method based on robust principal component sparse decomposition | |
Dharejo et al. | A deep hybrid neural network for single image dehazing via wavelet transform | |
Yadav et al. | A review on image fusion methodologies and applications | |
Sharma et al. | Satellite image contrast and resolution enhancement using discrete wavelet transform and singular value decomposition | |
Zhang et al. | Infrared and visible image fusion using joint convolution sparse coding | |
CN114298950A (en) | Infrared and visible light image fusion method based on improved GoDec algorithm | |
Xu et al. | AACNet: Asymmetric attention convolution network for hyperspectral image dehazing | |
Kaur et al. | A comparative study of various digital image fusion techniques: A review | |
Pai et al. | Medical color image enhancement using wavelet transform and contrast stretching technique | |
Indira et al. | Pixel based medical image fusion techniques using discrete wavelet transform and stationary wavelet transform | |
CN105096274A (en) | Infrared image noise reduction method based on non-subsampled contourlet domain mixed statistical model | |
Zhang et al. | Visible and infrared image fusion using convolutional dictionary learning with consensus auxiliary-auxiliary coupling | |
Amro et al. | General shearlet pansharpening method using Bayesian inference | |
CN114549379B (en) | Infrared and visible light image fusion method under non-downsampled shear wave transformation domain | |
Zhu et al. | Multifocus image fusion based on uniform discrete curvelet transform | |
Yuan et al. | An efficient method for traffic image denoising | |
Sangeetha et al. | Performance analysis of exemplar based image inpainting algorithms for natural scene image completion | |
CN112508829B (en) | Pan-sharpening method based on shear wave transformation |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20171208 Assignee: Guangxi Yanze Information Technology Co.,Ltd. Assignor: GUILIN University OF ELECTRONIC TECHNOLOGY Contract record no.: X2023980046249 Denomination of invention: A fusion algorithm for infrared and visible light images based on hybrid multiscale analysis Granted publication date: 20210622 License type: Common License Record date: 20231108 |
|
EE01 | Entry into force of recordation of patent licensing contract |