CN109767389A - Adaptive weighted double blind super-resolution reconstruction methods of norm remote sensing images based on local and non local joint priori - Google Patents
Adaptive weighted double blind super-resolution reconstruction methods of norm remote sensing images based on local and non local joint priori Download PDFInfo
- Publication number
- CN109767389A CN109767389A CN201910035555.2A CN201910035555A CN109767389A CN 109767389 A CN109767389 A CN 109767389A CN 201910035555 A CN201910035555 A CN 201910035555A CN 109767389 A CN109767389 A CN 109767389A
- Authority
- CN
- China
- Prior art keywords
- local
- resolution
- priori
- blind
- norm
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Image Processing (AREA)
Abstract
The invention discloses a kind of adaptive weighted double blind super-resolution reconstruction methods of norm remote sensing images based on local and non local joint priori.It mainly comprises the steps that in fuzzy kernel estimates subprocess using adaptive weighted double norm priori, with the fuzzy core and initial high-resolution image estimated；Using the fuzzy core of estimation and initial high-resolution image as the non-blind input for rebuilding subprocess；In non-blind reconstruction subprocess, the fuzzy core and initial high-resolution image of estimation estimate high-definition picture using local and non local joint priori and maximum a posteriori probability reconstruction model as known conditions.Using reconstructed results as new input high-definition picture, two above step is repeated until reaching maximum and rebuilds number, finally output is final reconstructed results.The method of the invention can be rebuild the high-definition picture of high quality under conditions of fuzzy core is unknown by low-resolution image, and the remote sensing images of reconstruction can be applied to the fields such as military, agricultural and the people's livelihood.
Description
Technical field
The present invention relates to image resolution ratio lift techniques, and in particular to it is a kind of based on it is local and it is non local joint priori from
It adapts to weight the blind super-resolution reconstruction method of double norm remote sensing images, belongs to field of image processing.
Background technique
High-resolution remote sensing images are widely used in the fields such as military affairs, agricultural, the people's livelihood.However, remote sensing images
Acquisition be easy the influence for being vibrated and moving, cause blur degradation, therefore the quality of remote sensing images may be not fully up to expectations.
Under normal conditions, image deterioration includes fuzzy and down-sampling.Therefore, in practical applications, it is directed to by image processing techniques
The quality for promoting to property the remote sensing images obtained, is necessary.In order to solve these problems, people are to image deblurring
It is conducted extensive research with increase resolution.This problem is studied herein, referred to as rebuilding blind super-resolution.Single width
Image super-resolution rebuilding technology be promoted image resolution ratio one of method, have be easily achieved, be at low cost, strong applicability
The features such as, it estimates corresponding high-definition picture by the single width low-resolution image observed.Due to same low-resolution image
Multiple and different high-definition pictures may be corresponded to, single image super-resolution rebuilding problem has serious pathosis.Therefore
It needs to constrain by image prior to obtain a stabilization, the estimation of reliable high-resolution.Current single image super-resolution
Method for reconstructing can substantially be divided into three classes: the super-resolution method based on interpolation, the super-resolution method based on reconstruction and be based on
The super-resolution method of study.These three types of methods have the characteristics that different, and such as the method based on interpolation is usually according only to interpolation kernel
Obtain interpolation image, less to consider fuzzy and noise influence, application range is relatively limited.Super-resolution based on reconstruction
Method sufficiently excavates the information of degraded image itself, usually can preferably inhibit artificial trace.Method based on study often has
Have it is faster execute speed, and can restoring image detail well.In addition, carrying out Super-resolution reconstruction under unknown fuzzy core
Build is that image restores the bigger challenge faced, as rebuilding blind super-resolution.Rebuilding blind super-resolution is generally divided into mould
Paste two subprocess of kernel estimates and non-rebuilding blind super-resolution.
Summary of the invention
The purpose of the present invention is organically combining two subprocess of fuzzy kernel estimates and non-blind super-resolution, and then construct
A kind of high efficiency, the blind method for reconstructing of high performance Remote Sensing Image Super Resolution.
Adaptive weighted double blind oversubscription of norm remote sensing images proposed by the present invention based on local and non local joint priori
It distinguishes method for reconstructing, mainly includes following operating procedure:
(1) in fuzzy kernel estimates subprocess, using adaptive weighted double norm priori, with the fuzzy core estimated and just
Beginning high-definition picture；
(2) using the fuzzy core of estimation and initial high-resolution image as the non-blind input for rebuilding subprocess；
(3) in non-blind reconstruction subprocess, the fuzzy core and initial high-resolution image of estimation utilize office as known conditions
Portion and non local joint priori and maximum a posteriori probability reconstruction model estimate high-definition picture.
(4) using the reconstructed results of step (3) as new input high-definition picture, step (3) and step are repeated
(4), number is built until reaching maximum, finally output is final reconstructed results.
Detailed description of the invention
Fig. 1 is that the present invention is based on adaptive weighted double blind super-resolution of norm remote sensing images of local and non local joint priori
The functional block diagram of method for reconstructing
Fig. 2 is initial high-resolution image and bicubic reconstruction image comparison diagram: where (a) (c) is fuzzy nuclear processes
Two times of reconstruction initial high-resolution images are obtained, (b) (d) is the result rebuild using two times of bicubic
Fig. 3 is the present invention and other methods fuzzy core estimated result comparison diagram: where (a) (e) (i) is realistic blur core
Image, (b) (f) (j) is the fuzzy core that the present invention estimates, (c) (g) (k) is the fuzzy core that deblurring control methods 1 is estimated, (d)
(h) (l) is the fuzzy core that deblurring control methods 2 is estimated
Fig. 4 is comparison diagram of the present invention with six kinds of methods to test image " mobilehomepark " two times of reconstructed results:
Wherein, (a) is input low-resolution image, (i) is original high-resolution image, and (b) (c) (d) (e) (f) (g) (h) is respectively
Control methods 1, control methods 2, control methods 3, control methods 4, control methods 5, Bicubic and reconstructed results of the invention
Specific embodiment
The present invention will be further explained below with reference to the attached drawings:
In Fig. 1, adaptive weighted double blind Super-resolution Reconstructions of norm remote sensing images based on local and non local joint priori
Method can specifically be divided into following steps:
(1) in fuzzy kernel estimates subprocess, using adaptive weighted double norm priori, with the fuzzy core estimated and just
Beginning high-definition picture；
(2) using the fuzzy core of estimation and initial high-resolution image as the non-blind input for rebuilding subprocess；
(3) in non-blind reconstruction subprocess, the fuzzy core and initial high-resolution image of estimation utilize office as known conditions
Portion and non local joint priori and maximum a posteriori probability reconstruction model estimate high-definition picture.
(4) using the reconstructed results of step (3) as new input high-definition picture, step (3) and step are repeated
(4), number is built until reaching maximum, finally output is final reconstructed results.
Specifically, in the step (1), we input low resolution blurred picture first, and building is general based on maximum a posteriori
The reconstruction framework of rate, using adaptive weighted double norm priori and convolution consistency priori as constraint condition in the frame,
In the fuzzy core of adaptive weighted double norm priori difference restrained split-flows and the high-definition picture of estimation, convolution consistency priori
The high-definition picture of restrained split-flow.As shown in formula (1):
λ indicates the parameter of first item de-blurred image convolution output；H is that the matrix of fuzzy core k indicates；D is that reduction is original
The down-sampling matrix of high-definition picture resolution ratio；X is original high-resolution image, and y is the low resolution fuzzy graph observed
Picture；αx, βx, αk, βkIt is regularization parameter；η is convolution consistency constraint parameter；It is convolution consistency constraint item, uses
In reduction pathosis, high-definition pictureIt can rebuild to obtain by existing super-resolution algorithms；By lpNorm and l2Norm building
Image prior itemWith fuzzy core priori itemAdaptive double norm weightings are collectively constituted
Priori, wherein weighting matrix W is for adaptively determining that image current region is non-edge or fringe region, and weighs according to this
L is emphasized again2Norm is for the smooth and noise suppressed effect of image non-edge and lpSharpening of the norm to fringe region
It acts on, wherein each w in WiIt is defined as formula (2):
It represents the local Non-smooth surface of a 5*5 image block centered on ith pixel.XiIt is the image block
Center pixel, ΩiIt is all pixels indexed set, X in the image blockijIt is XiNeighbor pixel in the position j.
By step (1), the fuzzy core k and initial high-resolution image x that can be estimated.
In the step (2), we rebuild high-definition picture by the non-blind reconstruction subprocess of super-resolution.Wherein in step
Suddenly the fuzzy core k and initial high-resolution image x of estimation obtained in (1) are as known input item, thus by former and later two
Subprocess combines, such as Fig. 1.
In the step (3), we construct the reconstruction framework based on maximum a posteriori probability, local and non local using joint
Priori as image prior, see formula (3):
Wherein JAHNLTVIt is non local image prior, JAGDIt is local image prior, ζ and θ are the above-mentioned two elder generations of balance
The regularization parameter tested.In this step, fuzzy core is known and initial high-resolution image is as iteration starting point, to reconstruct
High-definition picture.
In the step (4), we repeat step using the result of step (3) as new initial high-resolution image
Suddenly (3) and step (4).Number is rebuild until reaching the maximum of setting, reconstructed results are final output.
Validity in order to better illustrate the present invention has carried out initial high-resolution image comparative experiments respectively, obscures
Kernel estimates comparative experiments, and carried out on common test image " mobilehomepark " reconstruction of final high-definition picture
Comparative experiments.
Initial high-resolution image comparative experiments is as shown in Figure 2.Fig. 2 (a) and Fig. 2 (c) is by the fuzzy kernel estimates of the present invention
2 times of subprocess are rebuild obtained initial high-resolution image, and Fig. 2 (b) and Fig. 2 (d) are used to the low resolution of raw observation
The high-definition picture of 2 times of bicubic reconstructions.
Fuzzy kernel estimates comparative experiments is as shown in Figure 3.(a) (e) (i) is realistic blur core image, and (b) (f) (j) is this hair
The fuzzy core of bright estimation, (c) (g) (k) is the fuzzy core that deblurring control methods 1 is estimated, (d) (h) (l) is deblurring to analogy
The fuzzy core that method 2 is estimated.The algorithm of two kinds of comparisons are as follows:
The method that deblurring control methods 1:Xu et al. is proposed, bibliography " Xu L, Zheng S, Jia J.
“Unnatural l0 sparse representation for natural image deblurring,”Proceedings
of the IEEE conference on computer vision and pattern recognition.2013:1107-
1114.”。
The method that deblurring control methods 2:Shao et al. is proposed, bibliography " Shao W Z, Li H B, Elad M,
“Bi-l0-l2-norm regularization for blind motion deblurring,”Journal ofVisual
Communication and Image Representation,2015,33:42-59.”。
The reconstruction comparative experiments of final high-definition picture is as shown in Figure 4.(a) it is input low-resolution image, is (i) original
Beginning high-definition picture, (b) (c) (d) (e) (f) (g) (h) is respectively control methods 1, control methods 2, control methods 3, comparison
Method 4, control methods 5, Bicubic and reconstructed results of the invention.
The method that control methods 1:Shao et al. is proposed, bibliography " Shao W Z, Elad M, " Simple,
accurate,and robust nonparametric blind super-resolution,”International
Conference on Image and Graphics.Springer,Cham,2015:333-348.”。
Control methods 2: fuzzy kernel estimates subprocess uses deblurring control methods 2, non-blind reconstruction subprocess: Buades etc.
The method that people proposes, bibliography " Buades A, Coll B, Morel J M, " Image enhancement by non-
local reverse heat equation,”Preprint CMLA,2006,22:2006.”。
Control methods 3: fuzzy kernel estimates subprocess uses deblurring control methods 2, non-blind reconstruction subprocess: Ren et al.
The method of proposition, bibliography " Ren C, He X, Nguyen T Q, " Single image super-resolution via
adaptive high-dimensional non-local total variation and adaptive geometric
feature,”IEEE Transactions on Image Processing,2017,26(1):90-106.”。
Control methods 4: fuzzy kernel estimates subprocess uses deblurring control methods 1, non-blind reconstruction subprocess: Dong et al.
The method of proposition, bibliography " Dong W, Zhang L, Shi G, et al, " Nonlocally centralized sparse
representation for image restoration,”IEEE Transactions on Image Processing,
2013,22(4):1620-1630.”。
Control methods 5: fuzzy kernel estimates subprocess uses deblurring control methods 1, non-blind reconstruction subprocess: Buades etc.
The method that people proposes, bibliography " Buades A, Coll B, Morel J M, " Image enhancement by non-
local reverse heat equation,”Preprint CMLA,2006,22:2006.”。
The content of the reconstruction comparative experiments of final high-definition picture is as follows:
Bicubic is used respectively, and method 1, method 2, method 3, method 4, method 5 and the present invention are to by remote sensing test image
The low resolution blurred picture that library " UCMerced " simulation generates carries out 2 times of reconstructions.The fuzzy of low-resolution image degrades by eight
Kind fuzzy core is realized.Super-resolution rebuilding result to objectively evaluate parameter as shown in Table 1.Wherein objectively evaluate parameter PSNR
(Peak Signal to Noise Ratio), SSIM (Structure Similarity Index) are that value is bigger, are represented
Picture quality is better.Test of heuristics platform: the desk-top calculating of processor Inter Core i5CPU (3.3GHz) and memory 16G
Machine.
Table one
In the objective parameter shown in the table one, for eight kinds of different fuzzy cores present invention on remote sensing images test library
Highest PSNR, SSIM value is all achieved, the better quality of reconstructed results of the present invention is represent.
In conclusion reconstructed results of the invention have some superiority on subjective evaluation compared to control methods.Cause
This, the present invention is a kind of high performance single image super resolution ratio reconstruction method.
Claims (4)
1. special based on adaptive weighted double blind super-resolution reconstruction methods of norm remote sensing images of local and non local joint priori
Sign be the following steps are included:
Step 1: in fuzzy kernel estimates subprocess, using adaptive weighted double norm priori, with the fuzzy core estimated and just
Beginning high-definition picture；
Step 2: using the fuzzy core of estimation and initial high-resolution image as the non-blind input for rebuilding subprocess；
Step 3: in non-blind reconstruction subprocess, the fuzzy core and initial high-resolution image of estimation utilize office as known conditions
Portion and non local joint priori and maximum a posteriori probability reconstruction model estimate high-definition picture；
Step 4: using the reconstructed results of step 3 as new input high-definition picture, repeating step 3 and step 4,
Number is built until reaching maximum, finally output is final reconstructed results.
2. adaptive weighted double norm remote sensing images according to claim 1 based on local and non local joint priori are blind
Super-resolution reconstruction method, it is characterised in that adaptive weighted double norm priori described in step 1.
3. adaptive weighted double norm remote sensing images according to claim 1 based on local and non local joint priori are blind
Super-resolution reconstruction method, it is characterised in that part and non local joint described in step 3 based on initial high-resolution image
Priori.
4. adaptive weighted double norm remote sensing images based on local and non local joint priori described in -4 according to claim 1
Blind super-resolution reconstruction method, feature are being that fuzzy kernel estimates subprocess and super-resolution are combined using initial high-resolution image
The non-blind complete frame for rebuilding subprocess.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910035555.2A CN109767389B (en) | 2019-01-15 | 2019-01-15 | Self-adaptive weighted double-norm remote sensing image blind super-resolution reconstruction method based on local and non-local combined prior |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910035555.2A CN109767389B (en) | 2019-01-15 | 2019-01-15 | Self-adaptive weighted double-norm remote sensing image blind super-resolution reconstruction method based on local and non-local combined prior |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109767389A true CN109767389A (en) | 2019-05-17 |
| CN109767389B CN109767389B (en) | 2023-06-20 |
Family
ID=66454040
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910035555.2A Active CN109767389B (en) | 2019-01-15 | 2019-01-15 | Self-adaptive weighted double-norm remote sensing image blind super-resolution reconstruction method based on local and non-local combined prior |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109767389B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113096013A (en) * | 2021-03-31 | 2021-07-09 | 南京理工大学 | Blind image super-resolution reconstruction method and system based on imaging modeling and knowledge distillation |
| CN113627357A (en) * | 2021-08-13 | 2021-11-09 | 哈尔滨工业大学 | High-spatial-high-spectral-resolution intrinsic decomposition method and system for remote sensing image |
| CN114202459A (en) * | 2021-10-26 | 2022-03-18 | 北京工业大学 | Blind image super-resolution method based on depth prior |
| WO2023050258A1 (en) * | 2021-09-30 | 2023-04-06 | Baidu.Com Times Technology (Beijing) Co., Ltd. | Robust and efficient blind super-resolution using variational kernel autoencoder |
Citations (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2005055136A2 (en) * | 2003-11-26 | 2005-06-16 | Board Of Regents, The University Of Texas System | Method and apparatus for phase-sensitive magnetic resonance imaging |
| JP2008154237A (en) * | 1998-04-06 | 2008-07-03 | Seiko Epson Corp | Image data interpolation apparatus, computer for image data interpolation processing, image data interpolation method, and medium recorded with image data interpolation program |
| US20080158146A1 (en) * | 2007-01-03 | 2008-07-03 | Apple Computer, Inc. | Irregular input identification |
| US20100054595A1 (en) * | 2008-08-26 | 2010-03-04 | Microsoft Corporation | Automatic Image Straightening |
| WO2010050914A1 (en) * | 2008-10-31 | 2010-05-06 | Hewlett-Packard Development Company, L.P. | Method and system for enhancing image signals to alter the perception of depth by human viewers |
| US20110115934A1 (en) * | 2009-11-19 | 2011-05-19 | Sen Wang | Increasing image resolution using combined differential image |
| US20110122224A1 (en) * | 2009-11-20 | 2011-05-26 | Wang-He Lou | Adaptive compression of background image (acbi) based on segmentation of three dimentional objects |
| US20120300122A1 (en) * | 2011-05-26 | 2012-11-29 | Microsoft Corporation | Adaptive super resolution for video enhancement |
| US20140354886A1 (en) * | 2013-05-29 | 2014-12-04 | Yeda Research & Development Co. Ltd. | Device, system, and method of blind deblurring and blind super-resolution utilizing internal patch recurrence |
| CN104217405A (en) * | 2014-09-23 | 2014-12-17 | 闽江学院 | Salt-pepper noise filter method for image fusing local information and global information |
| US20150071567A1 (en) * | 2013-09-06 | 2015-03-12 | Kabushiki Kaisha Toshiba | Image processing device, image processing method and non-transitory computer readable medium |
| WO2015084966A2 (en) * | 2013-12-04 | 2015-06-11 | Razzor Technologies Inc. | Adaptive sharpening in image processing and display |
| WO2015180053A1 (en) * | 2014-05-28 | 2015-12-03 | 北京大学深圳研究生院 | Method and apparatus for rapidly reconstructing super-resolution image |
| CN105205783A (en) * | 2015-09-14 | 2015-12-30 | 河海大学 | SAR image blind super-resolution reestablishment method in combination with priori estimation |
| US9237257B1 (en) * | 2013-06-14 | 2016-01-12 | Xilinx, Inc. | Circuits for and methods of generating a digital image |
| CN105550989A (en) * | 2015-12-09 | 2016-05-04 | 西安电子科技大学 | Image super-resolution method based on nonlocal Gaussian process regression |
| CN105913392A (en) * | 2016-04-08 | 2016-08-31 | 西安电子科技大学昆山创新研究院 | Degraded image overall quality improving method in complex environment |
| CN106251297A (en) * | 2016-07-19 | 2016-12-21 | 四川大学 | A kind of estimation based on multiple image fuzzy core the rebuilding blind super-resolution algorithm of improvement |
| CN107025632A (en) * | 2017-04-13 | 2017-08-08 | 首都师范大学 | A kind of image super-resolution rebuilding method and system |
| CN107680040A (en) * | 2017-09-26 | 2018-02-09 | 北京空间飞行器总体设计部 | A kind of blind convolution super-resolution reconstruction method of multiframe based on bayesian criterion and device |
| CN109064402A (en) * | 2018-07-27 | 2018-12-21 | 四川大学 | Based on the single image super resolution ratio reconstruction method for enhancing non local total variation model priori |
-
2019
- 2019-01-15 CN CN201910035555.2A patent/CN109767389B/en active Active
Patent Citations (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008154237A (en) * | 1998-04-06 | 2008-07-03 | Seiko Epson Corp | Image data interpolation apparatus, computer for image data interpolation processing, image data interpolation method, and medium recorded with image data interpolation program |
| WO2005055136A2 (en) * | 2003-11-26 | 2005-06-16 | Board Of Regents, The University Of Texas System | Method and apparatus for phase-sensitive magnetic resonance imaging |
| US20080158146A1 (en) * | 2007-01-03 | 2008-07-03 | Apple Computer, Inc. | Irregular input identification |
| US20100054595A1 (en) * | 2008-08-26 | 2010-03-04 | Microsoft Corporation | Automatic Image Straightening |
| WO2010050914A1 (en) * | 2008-10-31 | 2010-05-06 | Hewlett-Packard Development Company, L.P. | Method and system for enhancing image signals to alter the perception of depth by human viewers |
| US20110115934A1 (en) * | 2009-11-19 | 2011-05-19 | Sen Wang | Increasing image resolution using combined differential image |
| US20110122224A1 (en) * | 2009-11-20 | 2011-05-26 | Wang-He Lou | Adaptive compression of background image (acbi) based on segmentation of three dimentional objects |
| US20120300122A1 (en) * | 2011-05-26 | 2012-11-29 | Microsoft Corporation | Adaptive super resolution for video enhancement |
| US20140354886A1 (en) * | 2013-05-29 | 2014-12-04 | Yeda Research & Development Co. Ltd. | Device, system, and method of blind deblurring and blind super-resolution utilizing internal patch recurrence |
| US9237257B1 (en) * | 2013-06-14 | 2016-01-12 | Xilinx, Inc. | Circuits for and methods of generating a digital image |
| US20150071567A1 (en) * | 2013-09-06 | 2015-03-12 | Kabushiki Kaisha Toshiba | Image processing device, image processing method and non-transitory computer readable medium |
| WO2015084966A2 (en) * | 2013-12-04 | 2015-06-11 | Razzor Technologies Inc. | Adaptive sharpening in image processing and display |
| WO2015180053A1 (en) * | 2014-05-28 | 2015-12-03 | 北京大学深圳研究生院 | Method and apparatus for rapidly reconstructing super-resolution image |
| CN104217405A (en) * | 2014-09-23 | 2014-12-17 | 闽江学院 | Salt-pepper noise filter method for image fusing local information and global information |
| CN105205783A (en) * | 2015-09-14 | 2015-12-30 | 河海大学 | SAR image blind super-resolution reestablishment method in combination with priori estimation |
| CN105550989A (en) * | 2015-12-09 | 2016-05-04 | 西安电子科技大学 | Image super-resolution method based on nonlocal Gaussian process regression |
| CN105913392A (en) * | 2016-04-08 | 2016-08-31 | 西安电子科技大学昆山创新研究院 | Degraded image overall quality improving method in complex environment |
| CN106251297A (en) * | 2016-07-19 | 2016-12-21 | 四川大学 | A kind of estimation based on multiple image fuzzy core the rebuilding blind super-resolution algorithm of improvement |
| CN107025632A (en) * | 2017-04-13 | 2017-08-08 | 首都师范大学 | A kind of image super-resolution rebuilding method and system |
| CN107680040A (en) * | 2017-09-26 | 2018-02-09 | 北京空间飞行器总体设计部 | A kind of blind convolution super-resolution reconstruction method of multiframe based on bayesian criterion and device |
| CN109064402A (en) * | 2018-07-27 | 2018-12-21 | 四川大学 | Based on the single image super resolution ratio reconstruction method for enhancing non local total variation model priori |
Non-Patent Citations (2)
| Title |
|---|
| 李滔等: "基于自适应双l_p-l_2范数的单幅模糊图像超分辨率盲重建", 《计算机应用》 * |
| 韩飞: "图像和视频的盲超分辨率算法研究", 《中国硕士学位论文全文数据库工程科技Ⅰ辑》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113096013A (en) * | 2021-03-31 | 2021-07-09 | 南京理工大学 | Blind image super-resolution reconstruction method and system based on imaging modeling and knowledge distillation |
| CN113627357A (en) * | 2021-08-13 | 2021-11-09 | 哈尔滨工业大学 | High-spatial-high-spectral-resolution intrinsic decomposition method and system for remote sensing image |
| CN113627357B (en) * | 2021-08-13 | 2022-07-08 | 哈尔滨工业大学 | High-spatial-high-spectral-resolution intrinsic decomposition method and system for remote sensing image |
| WO2023050258A1 (en) * | 2021-09-30 | 2023-04-06 | Baidu.Com Times Technology (Beijing) Co., Ltd. | Robust and efficient blind super-resolution using variational kernel autoencoder |
| CN114202459A (en) * | 2021-10-26 | 2022-03-18 | 北京工业大学 | Blind image super-resolution method based on depth prior |
| CN114202459B (en) * | 2021-10-26 | 2024-03-22 | 北京工业大学 | Blind image super-resolution method based on depth priori |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109767389B (en) | 2023-06-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wang et al. | Real-esrgan: Training real-world blind super-resolution with pure synthetic data | |
| Jin et al. | A flexible deep CNN framework for image restoration | |
| Zhang et al. | Beyond a gaussian denoiser: Residual learning of deep cnn for image denoising | |
| Huang et al. | Robust single-image super-resolution based on adaptive edge-preserving smoothing regularization | |
| Dong et al. | Nonlocally centralized sparse representation for image restoration | |
| Zhang et al. | Image super-resolution based on structure-modulated sparse representation | |
| Dong et al. | Centralized sparse representation for image restoration | |
| CN109767389A (en) | Adaptive weighted double blind super-resolution reconstruction methods of norm remote sensing images based on local and non local joint priori | |
| Min et al. | Blind deblurring via a novel recursive deep CNN improved by wavelet transform | |
| CN108460723B (en) | Bilateral total variation image super-resolution reconstruction method based on neighborhood similarity | |
| Ren et al. | Learning image profile enhancement and denoising statistics priors for single-image super-resolution | |
| Zhao et al. | Deep pyramid generative adversarial network with local and nonlocal similarity features for natural motion image deblurring | |
| Huang et al. | Learning deformable and attentive network for image restoration | |
| Kong et al. | Image reconstruction with predictive filter flow | |
| Ren et al. | Remote sensing image recovery via enhanced residual learning and dual-luminance scheme | |
| Wang et al. | An improved image blind deblurring based on dark channel prior | |
| Laroche et al. | Deep model-based super-resolution with non-uniform blur | |
| CN111986121A (en) | Based on Framellet l0Norm-constrained fuzzy image non-blind restoration method | |
| Du et al. | CNN-based image super-resolution and deblurring | |
| Ren et al. | Compressed image restoration via deep deblocker driven unified framework | |
| Xie et al. | Bidirectionally aligned sparse representation for single image super-resolution | |
| Xiang et al. | Learning spatio-temporal downsampling for effective video upscaling | |
| Esmaeilzehi et al. | DSegAN: A deep light-weight segmentation-based attention network for image restoration | |
| Pei et al. | Super-resolution of face images using weighted elastic net constrained sparse representation | |
| Nawaz et al. | Image super resolution by sparse linear regression and iterative back projection |
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 |