CN106920223A - A kind of small echo and rational rank partial differential joint image Enhancement Method - Google Patents

A kind of small echo and rational rank partial differential joint image Enhancement Method Download PDF

Info

Publication number
CN106920223A
CN106920223A CN201710150315.8A CN201710150315A CN106920223A CN 106920223 A CN106920223 A CN 106920223A CN 201710150315 A CN201710150315 A CN 201710150315A CN 106920223 A CN106920223 A CN 106920223A
Authority
CN
China
Prior art keywords
image
wavelet
rsqb
lsqb
information
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
Application number
CN201710150315.8A
Other languages
Chinese (zh)
Inventor
张勋
时延利
张宏瀚
严浙平
徐健
陈涛
周佳加
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Harbin Engineering University
Original Assignee
Harbin Engineering University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Harbin Engineering University filed Critical Harbin Engineering University
Priority to CN201710150315.8A priority Critical patent/CN106920223A/en
Publication of CN106920223A publication Critical patent/CN106920223A/en
Pending legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00Image enhancement or restoration
    • G06T5/10Image enhancement or restoration using non-spatial domain filtering
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/10Image acquisition modality
    • G06T2207/10048Infrared image
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/20Special algorithmic details
    • G06T2207/20048Transform domain processing
    • G06T2207/20064Wavelet transform [DWT]
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/20Special algorithmic details
    • G06T2207/20172Image enhancement details
    • G06T2207/20192Edge enhancement; Edge preservation
    • YGENERAL 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
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02TCLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
    • Y02T10/00Road transport of goods or passengers
    • Y02T10/10Internal combustion engine [ICE] based vehicles
    • Y02T10/40Engine management systems

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Image Processing (AREA)

Abstract

The present invention is to provide a kind of small echo and rational rank partial differential joint image Enhancement Method.The visible images and infrared image on UUV seas under acquisition surge environment;The improvement template in 3 directions of rational rank partial differential operator is obtained using the image enhaucament model based on rational rank partial differential;Wavelet decomposition is carried out respectively to infrared image and visible images, high and low frequency wavelet coefficient is obtained;Wavelet low frequency coefficient and high frequency coefficient are processed with rational rank partial differential operator template and 3 improvement templates in direction, for extracting image edge information;Then to image information wavelet inverse transformation, Image Reconstruction is carried out, obtains the visible images and infrared image on UUV seas under enhanced surge environment.The enhanced image of the method for the present invention reaches preferable enhancing effect in terms of image definition and medium, high frequency Edge texture details.

Description

A kind of small echo and rational rank partial differential joint image Enhancement Method
Technical field
The present invention relates to a kind of infrared and visual light imaging technology, especially a kind of image enchancing method.
Background technology
Rational rank partial differential algorithm for image enhancement can to a certain extent strengthen the high fdrequency component and smooth figure of image The low frequency component of picture, but it is not ideal enough to the Edge texture details enhancing effect of image medium, high frequency.
Document《Image enhaucament new model based on rational rank partial differential》(Sichuan University's journal (natural science edition), In January, 2016, the 1st phase of volume 53) fractional order and integer rank calculus theory are combined, brand-new rational rank differential is derived, So as to construct the image enhaucament model based on rational rank partial differential in spatial domain, new model has merged integer rank differential and has divided The number rank respective advantage of differential, compensate for respective deficiency, and enhancing model is realized using rational rank partial differential mask operator Numerical computations, the high fdrequency component and smoothed image low frequency component of the method enhancing image can obtain the increasing of consecutive variations to image Potent fruit, but it is undesirable to the grain details enhancing effect at image medium, high frequency edge.
Document《Small echo and fractional order differential joint image enhancing algorithm》(control engineering, in September, 2015, volume 22 the 5th Phase) combine the enhancing image medium, high frequency marginal information and fractional order differential algorithm for image enhancement of Wavelet image enhancing algorithm It is non-linear retain image low-frequency information advantage, but to strengthen image high fdrequency component and smoothed image low frequency component effect not Good, image can not obtain the enhancing effect of consecutive variations.
The content of the invention
Infrared image can be remarkably reinforced it is an object of the invention to provide one kind and visible images medium, high frequency Edge texture is thin The small echo and rational rank partial differential joint image Enhancement Method of section information.
The object of the present invention is achieved like this:
1st, obtain surge environment under UUV seas visible images and infrared image;
2nd, 3 directions of rational rank partial differential operator are obtained using the image enhaucament model based on rational rank partial differential Improvement template;
3rd, wavelet decomposition is carried out respectively to infrared image and visible images, obtains high and low frequency wavelet coefficient;
4th, wavelet low frequency coefficient and high frequency are processed with rational rank partial differential operator template and 3 improvement templates in direction Coefficient, for extracting image edge information;
5 and then to image information wavelet inverse transformation, Image Reconstruction is carried out, obtain UUV seas under enhanced surge environment Visible images and infrared image.
The present invention can also include:
1st, the improvement template in 3 directions is that using the time-frequency characteristic of wavelet algorithm, design is gone horizontal direction, goes to hang down Nogata to, go the improvement template in diagonally opposed 3 directions, going horizontal direction to improve template is used to extract on image level direction Vertical direction and diagonally opposed high-frequency information;Vertical direction template is gone for the horizontal direction extracted in image vertical direction and right Angular direction high-frequency information;Go diagonally opposed template for extract image it is diagonally opposed on horizontally and vertically high frequency believe Breath.
2nd, it is described that wavelet decomposition is carried out respectively to infrared image and visible images, obtain high and low frequency wavelet coefficient Step includes:
(1), obtaining wavelet decomposition formula according to Mallat algorithms is:
FormulaIt is the wavelet decomposition process of image low-frequency information, h [k] is Ideal low-pass filter, aj+1[k] is the discrete low frequency decomposed information of (j+1) level, is by upper level i.e. j-th stage low-frequency information aj [k] is obtained with h [k] convolutional calculation;FormulaIt is the small echo of image high-frequency information Decomposable process, g [k] is high-pass filter, dj+1[k] is (j+1) level discrete high frequency decomposed information, is by upper level i.e. j-th stage Low-frequency information aj[k] carries out convolutional calculation and obtains with g [k];
(2) one layer of wavelet decomposition, is carried out to image f (x, y), the wavelet coefficient da of low-frequency image L is obtained, horizontal direction is high Frequency image Hl, vertical direction high frequency imaging Hv, the wavelet coefficient dl of 3 direction high frequency imagings of diagonally opposed high frequency imaging Hd, dv、dd。
3rd, it is described processed with rational rank partial differential operator template and 3 improvement templates in direction wavelet low frequency coefficient and High frequency coefficient is specifically included:
(1) low-frequency image wavelet coefficient da is processed with rational rank partial differential operator template, is obtained
(2) the wavelet coefficient dl that horizontal direction improves template treatment horizontal direction high frequency imaging Hl is spent, is obtained
(3) the wavelet coefficient dv that vertical direction improves template treatment vertical high frequency image Hv is spent, is obtained
(4) the wavelet coefficient dd that diagonally opposed improvement template processes diagonal high frequency imaging Hd is spent, is obtained
4th, it is described to image information wavelet inverse transformation, Image Reconstruction is carried out, obtain UUV seas under enhanced surge environment The step of visible images and infrared image, includes:
(1) signal reconstruct formula is:
Wavelet information reconstruct is that wavelet information decomposes inverse process, j-th stage image low-frequency information aj[k] is by (j+1) level figure As low-frequency information aj+1[k+1] and image high-frequency information dj+1[k] is pressedFormula is reconstructed Arrive;For ideal low-pass filter, g [k] is ideal highpass filter to h [k];
(2) to wavelet coefficientInverse wavelet transform is carried out, i.e. wavelet information reconstruct finally gives side The enhanced new images of edge detailed information
In order to admirably achieve the enhancing effect of medium, high frequency Edge texture information, the present invention proposes small echo and rational rank Partial differential joint image strengthens algorithm.Time-frequency characteristic using wavelet algorithm strengthens image border texture information, image The high-frequency information and low-frequency information of multiple directions are broken down into after wavelet decomposition, using rational rank partial differential operator template and New improved template carries out the further extraction of marginal information to each component information after decomposition, then image information reconstruct, Obtain strengthening image.Enhanced image reaches preferably enhancing in terms of image definition and medium, high frequency Edge texture details Effect.By experiment simulation, visually observation enhancing image and the analysis for carrying out image quantitative target, it is known that, side of the invention Method is compared with rational rank partial differential algorithm for image enhancement, and method of the present invention image definition and medium, high frequency Edge texture are thin Enhancing effect in terms of section is better than single rational rank partial differential algorithm.
Small echo of the invention and rational rank partial differential joint image Enhancement Method not only remain rational rank partial differential The high fdrequency component of algorithm enhancing image and the effect of smoothed image low frequency component, and clearly enhance infrared image and visible ray Image medium, high frequency Edge texture detailed information, fully combines small echo with two methods of rational rank partial differential to image enhaucament Advantage, compensate for respective deficiency.
Brief description of the drawings
Fig. 1 is flow chart of the invention.
Fig. 2 a- Fig. 2 e are the method for the present invention and UUV seas under the surge environment of rational rank partial differential algorithm for image enhancement The infrared image enhancement effect contrast figure in face.Wherein Fig. 2 a are artwork;Fig. 2 b strengthen for p=1.6 rational rank partial differentials algorithm Image;Fig. 2 c are that p=1.6 the inventive method strengthens image;Fig. 2 d are that p=1.8 rational rank partial differentials algorithm strengthens image;Figure 2e is that p=1.8 the inventive method strengthens image.
Fig. 3 a- Fig. 3 e are the method for the present invention and UUV seas under rational rank partial differential algorithm for image enhancement surge environment Visible images enhancing effect comparison diagram.Wherein Fig. 3 a are artwork;Fig. 3 b strengthen for p=1.6 rational rank partial differentials algorithm Image;Fig. 3 c are that p=1.6 the inventive method strengthens image;Fig. 3 d are that p=1.8 rational rank partial differentials algorithm strengthens image;Figure 3e is that p=1.8 the inventive method strengthens image.
Fig. 4 a- Fig. 4 f are imitated for the infrared image enhancement on method of the present invention UUV seas under surge environment under different rank Fruit comparison diagram.Wherein Fig. 4 a are original image;Fig. 4 b are p=1.2;Fig. 4 c are p=1.4;Fig. 4 d are p=1.6;Fig. 4 e are p= 1.8;Fig. 4 f are p=2.0.
Fig. 5 is the visible images enhancing effect pair on method of the present invention UUV seas under surge environment under different rank Than figure.Wherein Fig. 5 a are original image;Fig. 5 b are p=1.2;Fig. 5 c are p=1.4;Fig. 5 d are p=1.6;Fig. 5 e are p=1.8; Fig. 5 f are p=2.0.
Specific embodiment
Illustrate below in conjunction with the accompanying drawings and the present invention be described in more detail, but protection scope of the present invention be not limited to it is following It is described.
As shown in figure 1, a kind of small echo and rational rank partial differential joint image enhancing algorithm, comprise the following steps:
1st, the improvement of rational rank partial differential operator is obtained using the image enhaucament new model based on rational rank partial differential Template, according to the time-frequency characteristic of wavelet algorithm, devise 3 directions (go horizontal direction, go vertical direction, go it is diagonally opposed) Template is improved, it is to further extract the vertical direction on image level direction and diagonally opposed to go horizontal direction to improve template High-frequency information;It is to further extract the horizontal direction in image vertical direction and diagonally opposed high frequency to go vertical direction template Information;Go diagonally opposed template and be in order to further extract image it is diagonally opposed on horizontally and vertically high frequency letter Breath.The template of rational rank partial differential operator is shown in Table 1.Rational rank partial differential operator improves template:Horizontal direction template is gone to see Table 2, goes vertical direction template to be shown in Table 3, goes diagonally opposed template to be shown in Table 4.
The template of the rational rank partial differential operator of table 1
Table 2 goes horizontal direction template
Table 3 goes vertical direction template
Table 4 goes diagonally opposed template
2nd, wavelet decomposition is carried out respectively to infrared image and visible images, obtains high and low frequency wavelet coefficient.
(1) know that wavelet decomposition formula is as follows according to Mallat algorithms:
(1) formula is the wavelet decomposition process of image low-frequency information, and h [k] is ideal low-pass filter, aj+1[k] is (j+ 1) the discrete low frequency decomposed information of level, is by upper level (j-th stage) low-frequency information aj[k] is obtained with h [k] convolutional calculation;(2) formula is The wavelet decomposition process of image high-frequency information, g [k] is high-pass filter, dj+1[k] is that (j+1) level discrete high frequency decomposes letter Breath, is by upper level (j-th stage) low-frequency information aj[k] carries out convolutional calculation and obtains with g [k].
(2) one layer of wavelet decomposition is carried out to image f (x, y), the wavelet coefficient da of low-frequency image L, 3 direction high frequencies is obtained The wavelet coefficient dl of image (horizontal direction high frequency imaging Hl, vertical direction high frequency imaging Hv, diagonally opposed high frequency imaging Hd), dv、dd。
3rd, wavelet low frequency coefficient and high frequency system are processed with rational rank partial differential operator template and new improved 3 templates Number, in order to extract image edge information.
(1) image low frequency profile, is preferably extracted, low-frequency image wavelet systems is processed with rational rank partial differential operator template Number da, obtains
(2), more preferably to extract vertical, diagonally opposed edge detail information, spend horizontal direction and improve template treatment level The wavelet coefficient dl of direction high frequency imaging Hl, obtains
(3), more preferably to extract level, diagonally opposed edge detail information, spend vertical direction and improve template treatment vertically The wavelet coefficient dv of high frequency imaging Hv, obtains
(4), more preferably to extract level, vertical direction edge detail information, diagonally opposed improvement template treatment is spent diagonal The wavelet coefficient dd of high frequency imaging Hd, obtains
4th, to image information wavelet inverse transformation, Image Reconstruction is carried out, obtains the visible of UUV seas under enhanced surge environment Light image and infrared image.
(1) signal reconstruct formula is:
Wavelet information reconstruct is that wavelet information decomposes inverse process, j-th stage image low-frequency information aj[k] is by (j+1) level figure As low-frequency information aj+1[k+1] and image high-frequency information dj+1[k] is obtained by the reconstruct of (6) formula;H [k] for ideal low-pass filter, G [k] is ideal highpass filter.
(2) to wavelet coefficientInverse wavelet transform, that is, wavelet information reconstruct are carried out, it is final to obtain To the new images that edge detail information is enhanced
In order to comprehensively contrast the image enhancement effects of the present invention and rational rank partial differential algorithm, the present invention is respectively adopted The infrared image and visible images on UUV seas carry out emulation experiment under surge environment, and image enhancement effects comparison diagram is to use The emulation that Matlab2014a is carried out.The image shot under surge environment is very fuzzy, it is desirable to obtain the good image of enhancing effect simultaneously It is not easy, Fig. 2 a- Fig. 2 e and Fig. 3 a- Fig. 3 e are the image enhancement effects comparison diagrams of the present invention and rational rank partial differential algorithm.
Then also emulation experiment has been done in match exponents change to the present invention, and the infrared figure on UUV seas under surge environment is respectively adopted Picture and visible images carry out emulation experiment, image enhancement effects of the present invention under different rank are compared see Fig. 4 a- Fig. 4 f, Fig. 5 a- Fig. 5 f.

Claims (5)

1. a kind of small echo and rational rank partial differential joint image Enhancement Method, it is characterized in that:
(1) obtain surge environment under UUV seas visible images and infrared image;
(2) 3 directions of rational rank partial differential operator are obtained using the image enhaucament model based on rational rank partial differential Improve template;
(3) wavelet decomposition is carried out respectively to infrared image and visible images, obtains high and low frequency wavelet coefficient;
(4) wavelet low frequency coefficient and high frequency system are processed with rational rank partial differential operator template and 3 improvement templates in direction Number, for extracting image edge information;
(5) and then to image information wavelet inverse transformation, Image Reconstruction is carried out, obtain UUV seas under enhanced surge environment can See light image and infrared image.
2. small echo according to claim 1 and rational rank partial differential joint image Enhancement Method, it is characterized in that:Described 3 The improvement template in individual direction is that using the time-frequency characteristic of wavelet algorithm, design is gone horizontal direction, goes vertical direction, goes diagonal side To 3 improvement templates in direction, going horizontal direction to improve template is used for the vertical direction extracted on image level direction and diagonal Direction high-frequency information;Vertical direction template is gone to believe for the horizontal direction extracted in image vertical direction and diagonally opposed high frequency Breath;Go diagonally opposed template for extract image it is diagonally opposed on horizontally and vertically high-frequency information.
3. small echo according to claim 2 and rational rank partial differential joint image Enhancement Method, it is characterized in that:It is described right Infrared image and visible images carry out wavelet decomposition respectively, include the step of obtain high and low frequency wavelet coefficient:
(1), obtaining wavelet decomposition formula according to Mallat algorithms is:
a j + 1 [ k ] = Σ n = - ∞ + ∞ a j [ n ] h [ n - 2 k ] = a j [ k ] * h [ k ]
d j + 1 [ k ] = Σ n = - ∞ + ∞ a j [ n ] g [ n - 2 k ] = a j [ k ] * h [ k ]
FormulaIt is the wavelet decomposition process of image low-frequency information, h [k] is preferable Low pass filter, aj+1[k] is the discrete low frequency decomposed information of (j+1) level, is by upper level i.e. j-th stage low-frequency information aj[k] with H [k] convolutional calculation is obtained;FormulaIt is the wavelet decomposition mistake of image high-frequency information Journey, g [k] is high-pass filter, dj+1[k] is (j+1) level discrete high frequency decomposed information, is believed by upper level i.e. j-th stage low frequency Breath aj[k] carries out convolutional calculation and obtains with g [k];
(2) one layer of wavelet decomposition, is carried out to image f (x, y), the wavelet coefficient da of low-frequency image L, horizontal direction high frequency figure is obtained As Hl, vertical direction high frequency imaging Hv, wavelet coefficient dl, dv, dd of Hd3 direction high frequency imaging of diagonally opposed high frequency imaging.
4. small echo according to claim 3 and rational rank partial differential joint image Enhancement Method, it is characterized in that the use Rational rank partial differential operator template and 3 improvement templates in direction are specifically wrapped processing wavelet low frequency coefficient and high frequency coefficient Include:
(1) low-frequency image wavelet coefficient da is processed with rational rank partial differential operator template, is obtained
(2) the wavelet coefficient dl that horizontal direction improves template treatment horizontal direction high frequency imaging Hl is spent, is obtained
d l ‾ ( x , y ) = Σ s = - a + a Σ t = - b + b w a ( s , t ) d l ( x + s , y + t )
(3) the wavelet coefficient dv that vertical direction improves template treatment vertical high frequency image Hv is spent, is obtained
d v ‾ ( x , y ) = Σ s = - a + a Σ t = - b + b w b ( s , t ) d v ( x + s , y + t )
(4) the wavelet coefficient dd that diagonally opposed improvement template processes diagonal high frequency imaging Hd is spent, is obtained
d d ‾ ( x , y ) = Σ s = - a + a Σ t = - b + b w c ( s , t ) d d ( x + s , y + t ) .
5. small echo according to claim 4 and rational rank partial differential joint image Enhancement Method, it is characterized in that described right Image information wavelet inverse transformation, carries out Image Reconstruction, obtains under enhanced surge environment the visible images on UUV seas and infrared The step of image, includes:
(1) signal reconstruct formula is:
a j [ k ] = Σ n = - ∞ + ∞ a j + 1 [ n ] h [ k - 2 n ] + Σ n = - ∞ + ∞ d j + 1 [ n ] g [ k - 2 n ] = a j + 1 [ k ] * h [ k ] + d j + 1 [ k ] * g [ k ]
Wavelet information reconstruct is that wavelet information decomposes inverse process, j-th stage image low-frequency information aj[k] is by (j+1) level image low frequency Information aj+1[k+1] and image high-frequency information dj+1[k] is pressedFormula reconstruct is obtained;h[k] For ideal low-pass filter, g [k] is ideal highpass filter;
(2) to wavelet coefficientInverse wavelet transform is carried out, i.e. wavelet information reconstruct finally gives edge thin The enhanced new images of section information
CN201710150315.8A 2017-03-14 2017-03-14 A kind of small echo and rational rank partial differential joint image Enhancement Method Pending CN106920223A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710150315.8A CN106920223A (en) 2017-03-14 2017-03-14 A kind of small echo and rational rank partial differential joint image Enhancement Method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710150315.8A CN106920223A (en) 2017-03-14 2017-03-14 A kind of small echo and rational rank partial differential joint image Enhancement Method

Publications (1)

Publication Number Publication Date
CN106920223A true CN106920223A (en) 2017-07-04

Family

ID=59461362

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710150315.8A Pending CN106920223A (en) 2017-03-14 2017-03-14 A kind of small echo and rational rank partial differential joint image Enhancement Method

Country Status (1)

Country Link
CN (1) CN106920223A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112435195A (en) * 2020-12-02 2021-03-02 湖南优象科技有限公司 Image enhancement method and system based on adaptive fractional order differential
WO2021102644A1 (en) * 2019-11-25 2021-06-03 中国科学院深圳先进技术研究院 Image enhancement method and apparatus, and terminal device
CN112435195B (en) * 2020-12-02 2024-06-04 湖南优象科技有限公司 Image enhancement method and system based on self-adaptive fractional order differentiation

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102646272A (en) * 2012-02-23 2012-08-22 南京信息工程大学 Wavelet meteorological satellite cloud image merging method based on local variance and weighing combination
CN105931201A (en) * 2016-04-20 2016-09-07 北京航空航天大学 Image subjective visual effect enhancing method based on wavelet transformation
CN106097274A (en) * 2016-06-20 2016-11-09 陕西理工学院 A kind of adaptive fractional rank differential algorithm for image enhancement

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102646272A (en) * 2012-02-23 2012-08-22 南京信息工程大学 Wavelet meteorological satellite cloud image merging method based on local variance and weighing combination
CN105931201A (en) * 2016-04-20 2016-09-07 北京航空航天大学 Image subjective visual effect enhancing method based on wavelet transformation
CN106097274A (en) * 2016-06-20 2016-11-09 陕西理工学院 A kind of adaptive fractional rank differential algorithm for image enhancement

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
蒋伟 等: "基于有理数阶偏微分的图像增强新模型", 《四川大学学报(自然科学版)》 *
陈莉: "小波和分数阶微分联合图像增强算法", 《控制工程》 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2021102644A1 (en) * 2019-11-25 2021-06-03 中国科学院深圳先进技术研究院 Image enhancement method and apparatus, and terminal device
CN112435195A (en) * 2020-12-02 2021-03-02 湖南优象科技有限公司 Image enhancement method and system based on adaptive fractional order differential
CN112435195B (en) * 2020-12-02 2024-06-04 湖南优象科技有限公司 Image enhancement method and system based on self-adaptive fractional order differentiation

Similar Documents

Publication Publication Date Title
Cui et al. Detail preserved fusion of visible and infrared images using regional saliency extraction and multi-scale image decomposition
JP7260243B2 (en) Noisy image upscaling method and apparatus for noisy image upscaling
CN103186888B (en) A kind of method and device of removal CT picture noises
CN109685728B (en) Digital image processing method based on local time-frequency domain transformation
CN106204449A (en) A kind of single image super resolution ratio reconstruction method based on symmetrical degree of depth network
CN101504766B (en) Image amalgamation method based on mixed multi-resolution decomposition
CN106952228A (en) The super resolution ratio reconstruction method of single image based on the non local self-similarity of image
CN104182939B (en) Medical image detail enhancement method
CN109509160A (en) A kind of remote sensing image fusion method by different level using layer-by-layer iteration super-resolution
CN103208097B (en) Filtering method is worked in coordination with in the principal component analysis of the multi-direction morphosis grouping of image
CN110322404B (en) Image enhancement method and system
CN112163994B (en) Multi-scale medical image fusion method based on convolutional neural network
CN110211084B (en) Image multi-resolution reconstruction method based on weight wavelet transform
CN106204482A (en) Based on the mixed noise minimizing technology that weighting is sparse
CN106600564A (en) Novel image enhancement method
CN106097274A (en) A kind of adaptive fractional rank differential algorithm for image enhancement
CN104732498B (en) A kind of thresholded image denoising method based on non-downsampling Contourlet conversion
CN107895356A (en) A kind of near-infrared image Enhancement Method based on steerable pyramid
CN112785539A (en) Multi-focus image fusion method based on image adaptive decomposition and parameter adaptive
CN111815550A (en) Infrared and visible light image fusion method based on gray level co-occurrence matrix
CN103745443B (en) The method and apparatus for improving picture quality
Zhang et al. A novel method of medical image enhancement based on wavelet decomposition
CN106920223A (en) A kind of small echo and rational rank partial differential joint image Enhancement Method
CN104809714A (en) Image fusion method based on multi-morphological sparse representation
CN110175959B (en) Typhoon cloud picture enhancement method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication

Application publication date: 20170704

RJ01 Rejection of invention patent application after publication