CN106097274A - A kind of adaptive fractional rank differential algorithm for image enhancement - Google Patents
A kind of adaptive fractional rank differential algorithm for image enhancement Download PDFInfo
- Publication number
- CN106097274A CN106097274A CN201610443470.4A CN201610443470A CN106097274A CN 106097274 A CN106097274 A CN 106097274A CN 201610443470 A CN201610443470 A CN 201610443470A CN 106097274 A CN106097274 A CN 106097274A
- Authority
- CN
- China
- Prior art keywords
- image
- high frequency
- frequency
- algorithm
- differential
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
-
- 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
- G06T2207/20004—Adaptive image processing
-
- 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 Analysis (AREA)
Abstract
The invention discloses a kind of adaptive fractional rank differential algorithm for image enhancement, comprise the following steps: calculate image complexity according to differential box counting thought;The sublevel number of fractional order differential is calculated according to image complexity;Wavelet decomposition low-frequency image using fractional order differential template process and strengthens extraction detail of the high frequency, enhanced image retains low-frequency information;Horizontal direction high frequency detail image is processed;Wavelet decomposition vertical direction high frequency detail image processes;High frequency detail image diagonally opposed to wavelet decomposition processes;Low frequency after processing, level, vertical, diagonal angle wavelet decomposition image carry out wavelet inverse transformation and obtain strengthening image.The algorithm of the present invention, while retaining image low-frequency information, strengthens and extracts high-frequency information, and determine algorithm point order parameter according to image complexity self adaptation, is effectively guaranteed the optimum efficiency that algorithm strengthens.
Description
Technical field
The present invention relates to image algorithm, be specifically related to a kind of adaptive fractional rank differential algorithm for image enhancement.
Background technology
Fractional order differential is one of branch of mathematical analysis, in recent years many scholars to this theory in image procossing
Research is made in application, and current total thinking is to utilize the dependency of pixel and neighbor thereof, utilizes the multiple dimensioned structure can be real
Existing improvement template improves the reinforced effects of image border texture information, and its processing procedure is all carried out in spatial domain.
Existing fractional order differential algorithm can not provide optimal fractional order according to different feature of image and carry out image
Processing, these algorithms simply demonstrate sublevel number when being between 0-1, use fractional order differential algorithm can realize increasing to image
By force, along with the increase of sublevel number, the marginal information of extraction is the most, but it practice, sublevel number is more than after certain numerical value, the limit of extraction
Edge detailed information is pseudo-edge, it is therefore necessary to determine optimal sublevel number according to feature of image, uses optimal sublevel number to carry out
The enhancing of image.
Existing fractional order differential algorithm, be based primarily upon spatial domain and use fractional order differential template to process image, in order to extract
More edge detail information, reaches the target that better image strengthens.Image wavelet is decomposed by the present invention, for wavelet decomposition
Rear image is broken down into level, the principle of high-frequency information vertical, diagonally opposed, devises three processing template of correspondence, enters one
Step extracts the vertical direction of horizontal direction high frequency imaging, diagonally opposed high-frequency information, the level of vertical direction frequency image information,
Diagonally opposed high-frequency information, the level of diagonally opposed frequency image information, vertical direction high-frequency information, more image can be obtained
Marginal information, improves image enhancement effects.
When the existing method utilizing small echo and fractional order differential to combine processes image, the value of sublevel number is still 0-1, no
Optimal sublevel numerical value can be determined according to feature of image.
Summary of the invention
Present invention is primarily targeted at a kind of adaptive fractional rank of offer differential algorithm for image enhancement, including following step
Rapid:
S1, calculates image complexity according to differential box counting thought;
S2, calculates the sublevel number of fractional order differential according to image complexity;
S3, carries out wavelet transformation to image, and picture breakdown becomes low frequency, horizontal direction high frequency, vertical direction high frequency, diagonal angle
Direction high frequency time, frequency division solution image;
S4, uses wavelet decomposition low-frequency image fractional order differential template to process and strengthens extraction detail of the high frequency, strengthen
After image retain low-frequency information;In fractional order differential template, the value of sublevel number is determined by second;
S5, processes horizontal direction high frequency detail image, further enhances, extracts vertical, diagonal angle high frequency detail letter
Breath;In fractional order differential template, the value of sublevel number is determined by second;
S6, wavelet decomposition vertical direction high frequency detail image processes, and further enhances, extraction level, diagonally opposed
Detail of the high frequency;In fractional order differential template, the value of sublevel number is determined by second;
S7, high frequency detail image diagonally opposed to wavelet decomposition use create template, further enhance, extraction level,
Vertical direction detail of the high frequency;In fractional order differential template, the value of sublevel number is determined by second;
S8, low frequency after processing, level, vertical, diagonal angle wavelet decomposition image carry out wavelet inverse transformation and obtain enhancing figure
Picture, this image enhances high-frequency information while retaining low-frequency information.
Further, described step S1 particularly as follows:
M × M image is divided into L × L sub-block, wherein 1 < L < M/2, sub-block is placed the box that size is L × L × L'
Son, L' is the height of sub-block, and in order to fractal dimension D being limited between [2,3], L' meets [G/L']=[M/L], G and represents ash
Degree level sum;Assume in (i, j) vertical direction box numbering c that in individual grid, minimum gradation value is corresponding, maximum gradation value pair
Answer vertical direction box numbering d, then the box number dimension that in this grid, image is corresponding: nr(i, j)=d-c+1;Cover whole figure
The box number of picture isKnown fractal dimension computing formula is Substitute into NrAfter
To values of fractal dimension D.
Further, described step S2 is particularly as follows: use the method for V=D-2 to confirm the sublevel of fractional order differential algorithm
Number V.
Advantages of the present invention:
The algorithm of the present invention combines wavelet-decomposing method and carries out fractional order differential algorithm process image, and multiple according to image
Miscellaneous degree self adaptation determines algorithm point order parameter, is effectively guaranteed the optimum efficiency that algorithm strengthens.
In addition to objects, features and advantages described above, the present invention also has other objects, features and advantages.
Below with reference to figure, the present invention is further detailed explanation.
Accompanying drawing explanation
The accompanying drawing of the part constituting the application is used for providing a further understanding of the present invention, and the present invention's is schematic real
Execute example and illustrate for explaining the present invention, being not intended that inappropriate limitation of the present invention.
Fig. 1 is the adaptive fractional rank differential map image intensifying algorithm flow chart of the present invention;
Fig. 2 is lena.bmp image artwork and with at the adaptive fractional rank differential map image intensifying algorithm simulating of the present invention
The design sketch of reason.
Detailed description of the invention
In order to make the purpose of the present invention, technical scheme and advantage clearer, below in conjunction with drawings and Examples, right
The present invention is further elaborated.Should be appreciated that specific embodiment described herein only in order to explain the present invention, and
It is not used in the restriction present invention.
With reference to Fig. 1, a kind of adaptive fractional rank differential algorithm for image enhancement as shown in Figure 1, comprise the following steps:
S1, calculates image complexity according to differential box counting thought;
S2, calculates the sublevel number of fractional order differential according to image complexity;
S3, carries out wavelet transformation to image, and picture breakdown becomes low frequency, horizontal direction high frequency, vertical direction high frequency, diagonal angle
During the high frequency of direction, frequency division solution image;
S4, uses wavelet decomposition low-frequency image fractional order differential template to process and strengthens extraction detail of the high frequency, strengthen
After image retain low-frequency information;In fractional order differential template, the value of sublevel number is determined by second;The concrete template 1 that uses:
(v<sup>2</sup>-v)/2 | 0 | (v<sup>2</sup>-v)/2 | 0 | (v<sup>2</sup>-v)/2 |
0 | -v | -v | -v | 0 |
(v<sup>2</sup>-v)/2 | -v | 8 | -v | (v<sup>2</sup>-v)/2 |
0 | -v | -v | -v | 0 |
(v<sup>2</sup>-v)/2 | 0 | (v<sup>2</sup>-v)/2 | 0 | (v<sup>2</sup>-v)/2 |
S5, processes horizontal direction high frequency detail image, further enhances, extracts vertical, diagonal angle high frequency detail letter
Breath;In fractional order differential template, the value of sublevel number is determined by second;The concrete template 2 that uses:
(v<sup>2</sup>-v)/2 | 0 | (v<sup>2</sup>-v)/2 | 0 | (v<sup>2</sup>-v)/2 |
0 | -v | -v | -v | 0 |
0 | 0 | 6 | 0 | 0 |
0 | -v | -v | -v | 0 |
(v<sup>2</sup>-v)/2 | 0 | (v<sup>2</sup>-v)/2 | 0 | (v<sup>2</sup>-v)/2 |
S6, wavelet decomposition vertical direction high frequency detail image processes, and further enhances, extraction level, diagonally opposed
Detail of the high frequency.In fractional order differential template, the value of sublevel number is determined by second;The concrete template 3 that uses:
(v<sup>2</sup>-v)/2 | 0 | 0 | 0 | (v<sup>2</sup>-v)/2 |
0 | -v | 0 | -v | 0 |
(v<sup>2</sup>-v)/2 | -v | 6 | -v | (v<sup>2</sup>-v)/2 |
0 | -v | 0 | -v | 0 |
(v<sup>2</sup>-v)/2 | 0 | 0 | 0 | (v<sup>2</sup>-v)/2 |
S7, high frequency detail image diagonally opposed to wavelet decomposition use create template, further enhance, extraction level,
Vertical direction detail of the high frequency;In fractional order differential template, the value of sublevel number is determined by second;Concrete employing template 4
0 | 0 | (v<sup>2</sup>-v)/2 | 0 | 0 |
0 | 0 | -v | 0 | 0 |
(v<sup>2</sup>-v)/2 | -v | 4 | -v | (v<sup>2</sup>-v)/2 |
0 | 0 | -v | 0 | 0 |
0 | 0 | (v<sup>2</sup>-v)/8 | 0 | 0 |
S8, low frequency after processing, level, vertical, diagonal angle wavelet decomposition image carry out wavelet inverse transformation and obtain enhancing figure
Picture, this image enhances high-frequency information while retaining low-frequency information.
Described step S1 particularly as follows:
M × M image is divided into L × L sub-block, wherein 1 < L < M/2, sub-block is placed the box that size is L × L × L'
Son, L' is the height of sub-block, and in order to fractal dimension D being limited between [2,3], L' meets [G/L']=[M/L], G and represents ash
Degree level sum.Assume in (i, j) vertical direction box numbering c that in individual grid, minimum gradation value is corresponding, maximum gradation value pair
Answer vertical direction box numbering d, then the box number dimension that in this grid, image is corresponding: nr(i, j)=d-c+1.Cover whole figure
The box number of picture isKnown fractal dimension computing formula is Substitute into NrAfter
To values of fractal dimension D.
Described step S2 is particularly as follows: use the method for V=D-2 to confirm the sublevel number V of fractional order differential algorithm.Wherein D begins
It is distributed across the mark between [2,3] eventually.
The principle of the algorithm of the present invention:
1) utilizing difference dimension theory to determine image complexity, the differential box counting computational theory fractal according to image calculates
Go out the values of fractal dimension that can characterize image complexity, set up the mathematics pass of values of fractal dimension and fractional order differential algorithm sublevel number
System, determines sublevel number.
2) by image wavelet decompose, image be broken down into four parts: L be level, vertical direction be all low-frequency image, HL is
Horizontal direction high frequency imaging, LH is vertical direction high frequency, and HH is focusing direction high frequency.
LL | HL |
LH | HH |
Owing to wavelet transformation has time-frequency feature, image information is broken down into low-frequency information, level side after wavelet transformation
To high-frequency information, vertical direction high-frequency information, diagonally opposed high-frequency information, has directive feature;In order to extract further
High-frequency information, the feature for the high-frequency information in after wavelet decomposition 3 directions devises template 2, template 3, template 4.Template 2 is gone
Fall horizontal dimension coefficients, remained vertical direction, diagonally opposed coefficients, it is therefore an objective to extract horizontal direction high frequency further
The vertical direction of image, diagonally opposed high-frequency information (template 2 is used for processing horizontal direction high frequency imaging after wavelet decomposition);Template
3 eliminate vertical dimension coefficients, remain horizontal direction, diagonally opposed coefficients, it is therefore an objective to extract vertical direction further
The level of frequency image information, diagonally opposed high-frequency information (template 3 is used for processing vertical direction high frequency imaging after wavelet decomposition);
Template 4 eliminates diagonally opposed coefficient, remains horizontal direction, vertical direction coefficients, it is therefore an objective to extract diagonal angle further
(template 4 is diagonally opposed high frequency figure after being used for processing wavelet decomposition for the level of direction frequency image information, vertical direction high-frequency information
Picture).Coefficient inverse wavelet transform after finally processing, reconstructs image.The image of reconstruct is while non-linear reservation low-frequency information
High degree enhance detail of the high frequency.Compared with single fractional order differential image enchancing method, the algorithm of the present invention
Utilize the time-frequency characteristic of wavelet transformation, multi-resolution characteristics, use fractional order micro-on the basis of image carries out wavelet transformation
Point algorithm process is low, high-frequency wavelet coefficient, and the reinforced effects of its Edge texture detail of the high frequency is better than single spatial domain mark
Rank differential algorithm.
The experiment simulation of the present invention:
The present invention program is used to process lena.bmp image:
1) the difference dimension value D=2.4641 of lena.bmp image is calculated according to difference dimension theory.
2) according to the mapping relations of difference dimension value Yu sublevel number, optimal sublevel number v=0.4641 is drawn.
3) method combined processes image zooming-out image to use small echo and fractional order differential to want under this optimal sublevel number
Marginal information.
4) sublevel number value is edge image during 0.1-0.9, with the edge image pair during optimal sublevel number v=0.4641
Ratio, the edge image obtained under optimal sublevel number is the most abundant, and has minimum pseudo-edge.
With reference to Fig. 2, as shown in Figure 2:
1) as shown in (e) therein, the image edge information obtained under optimal sublevel number v=0.4641 is the most clear,
Best results;
2) as shown in (b) therein, (c), (d): when the value of sublevel number v is less than 4, the marginal information of extraction is less,
Marginal information is the most clear.
3) as shown in (f) therein, (h), (i), (j), after the value of sublevel number v is more than 5, edge image is extracted to obtain
In occur in that pseudo-edge, and along with the increase of sublevel number, the edge image pseudo-edge of extraction the most but comes the most.
4) as shown in wherein (k)-(m): strengthen image and enhance high-frequency information, v=while retaining low-frequency information
Enhancing image (l) obtained when 0.4641 is bigger than enhancing image (k) brightness that obtains during v=0.2, apparent, but during v=0.9,
Enhancing image (m), owing to being extracted pseudo-edge, thickens.
The present invention is according to the Computation schema of the fractal middle difference dimension box of image, it is proposed that true according to image complexity self adaptation
Determine differential fractional value to the method realizing image enhaucament.Being first depending on that the fractal differential box counting computational theory of image calculates can
To characterize the difference dimension value of image complexity, resettle the mathematics pass of difference dimension value and fractional order differential algorithm sublevel number
System, determines sublevel number, under this fractional value, and the method that application wavelet transformation is combined with fractional order differential, it is achieved according to image
The adaptive image enhancement of complexity.
Test result indicate that: this algorithm, while retaining image low-frequency information, strengthens and extract high-frequency information, and
Determine algorithm point order parameter according to image complexity self adaptation, be effectively guaranteed the optimum efficiency that algorithm strengthens.
The foregoing is only presently preferred embodiments of the present invention, not in order to limit the present invention, all spirit in the present invention and
Within principle, any modification, equivalent substitution and improvement etc. made, should be included within the scope of the present invention.
Claims (3)
1. an adaptive fractional rank differential algorithm for image enhancement, it is characterised in that comprise the following steps:
S1, calculates image complexity according to differential box counting thought;
S2, calculates the sublevel number of fractional order differential according to image complexity;
S3, carries out wavelet transformation to image, and picture breakdown becomes low frequency, horizontal direction high frequency, vertical direction high frequency, diagonally opposed
During high frequency, frequency division solution image;
S4, uses wavelet decomposition low-frequency image fractional order differential template to process and strengthens extraction detail of the high frequency, enhanced
Image retains low-frequency information;In fractional order differential template, the value of sublevel number is determined by second;
S5, processes horizontal direction high frequency detail image, further enhances, extracts vertical, diagonal angle detail of the high frequency;Point
In the differential template of number rank, the value of sublevel number is determined by second;
S6, wavelet decomposition vertical direction high frequency detail image processes, and further enhances, extraction level, diagonally opposed high frequency
Detailed information;In fractional order differential template, the value of sublevel number is determined by second;
S7, high frequency detail image diagonally opposed to wavelet decomposition uses the template created, and further enhances, extraction level, vertical
Direction detail of the high frequency;In fractional order differential template, the value of sublevel number is determined by second;
S8, low frequency after processing, level, vertical, diagonal angle wavelet decomposition image carry out wavelet inverse transformation and obtain strengthening image, should
Image enhances high-frequency information while retaining low-frequency information.
Adaptive fractional rank the most according to claim 1 differential algorithm for image enhancement, it is characterised in that described step S1 has
Body is:
M × M image is divided into L × L sub-block, wherein 1 < L < M/2, sub-block is placed the box that size is L × L × L', L'
Being the height of sub-block, in order to fractal dimension D being limited between [2,3], L' meets [G/L']=[M/L], and G represents that gray level is total
Number;Assume that (i, j) vertical direction box numbering c that in individual grid, minimum gradation value is corresponding, maximum gradation value is corresponding vertical the
Direction box numbering d, then the box number dimension that in this grid, image is corresponding: nr(i, j)=d-c+1;Cover the box of whole image
Subnumber isKnown fractal dimension computing formula isSubstitute into NrAfter divided
Shape dimension D.
Adaptive fractional rank the most according to claim 1 differential algorithm for image enhancement, it is characterised in that described step S2 has
Body is: use the method for V=D-2 to confirm the sublevel number V of fractional order differential algorithm.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610443470.4A CN106097274A (en) | 2016-06-20 | 2016-06-20 | A kind of adaptive fractional rank differential algorithm for image enhancement |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610443470.4A CN106097274A (en) | 2016-06-20 | 2016-06-20 | A kind of adaptive fractional rank differential algorithm for image enhancement |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106097274A true CN106097274A (en) | 2016-11-09 |
Family
ID=57238108
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610443470.4A Pending CN106097274A (en) | 2016-06-20 | 2016-06-20 | A kind of adaptive fractional rank differential algorithm for image enhancement |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106097274A (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106920223A (en) * | 2017-03-14 | 2017-07-04 | 哈尔滨工程大学 | A kind of small echo and rational rank partial differential joint image Enhancement Method |
CN107358585A (en) * | 2017-06-30 | 2017-11-17 | 西安理工大学 | Misty Image Enhancement Method based on fractional order differential and dark primary priori |
CN108036255A (en) * | 2017-12-04 | 2018-05-15 | 余姚市荣事特电子有限公司 | A kind of emergency light |
CN108226573A (en) * | 2017-12-29 | 2018-06-29 | 国网冀北电力有限公司张家口供电公司 | A kind of organic external insulation Analysis of Surface Topography method and device |
CN109636745A (en) * | 2018-11-28 | 2019-04-16 | 陕西理工大学 | Best rank image enchancing method based on fractional order differential algorithm for image enhancement |
CN110232670A (en) * | 2019-06-19 | 2019-09-13 | 重庆大学 | A method of the image visual effect enhancing based on low-and high-frequency separation |
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 |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140233820A1 (en) * | 2012-11-01 | 2014-08-21 | Virginia Commonweath University | Segmentation and Fracture Detection in CT Images |
CN104599260A (en) * | 2015-02-02 | 2015-05-06 | 天津三英精密仪器有限公司 | X-ray image enhancement method based on dual-energy spectrum and wavelet fusion |
CN104881847A (en) * | 2015-04-17 | 2015-09-02 | 广西科技大学 | Match video image enhancement method based on wavelet analysis and pseudo-color processing |
-
2016
- 2016-06-20 CN CN201610443470.4A patent/CN106097274A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140233820A1 (en) * | 2012-11-01 | 2014-08-21 | Virginia Commonweath University | Segmentation and Fracture Detection in CT Images |
CN104599260A (en) * | 2015-02-02 | 2015-05-06 | 天津三英精密仪器有限公司 | X-ray image enhancement method based on dual-energy spectrum and wavelet fusion |
CN104881847A (en) * | 2015-04-17 | 2015-09-02 | 广西科技大学 | Match video image enhancement method based on wavelet analysis and pseudo-color processing |
Non-Patent Citations (3)
Title |
---|
张艳珠 等: "分数阶微分增强的脑部MRI图像边缘检测", 《沈阳理工大学学报》 * |
李钊 等: "分形维数计算的流水线优化方法研究", 《仪器仪表学报》 * |
陈莉: "小波和分数阶微分联合图像增强算法", 《控制工程》 * |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106920223A (en) * | 2017-03-14 | 2017-07-04 | 哈尔滨工程大学 | A kind of small echo and rational rank partial differential joint image Enhancement Method |
CN107358585A (en) * | 2017-06-30 | 2017-11-17 | 西安理工大学 | Misty Image Enhancement Method based on fractional order differential and dark primary priori |
CN107358585B (en) * | 2017-06-30 | 2020-08-18 | 西安理工大学 | Foggy day image enhancement method based on fractional order differential and dark channel prior |
CN108036255A (en) * | 2017-12-04 | 2018-05-15 | 余姚市荣事特电子有限公司 | A kind of emergency light |
CN108226573A (en) * | 2017-12-29 | 2018-06-29 | 国网冀北电力有限公司张家口供电公司 | A kind of organic external insulation Analysis of Surface Topography method and device |
CN108226573B (en) * | 2017-12-29 | 2021-01-08 | 国网冀北电力有限公司张家口供电公司 | Fractal theory-based organic external insulation surface morphology analysis method and device |
CN109636745A (en) * | 2018-11-28 | 2019-04-16 | 陕西理工大学 | Best rank image enchancing method based on fractional order differential algorithm for image enhancement |
CN109636745B (en) * | 2018-11-28 | 2023-03-14 | 陕西理工大学 | Optimal order image enhancement method based on fractional order differential image enhancement algorithm |
CN110232670A (en) * | 2019-06-19 | 2019-09-13 | 重庆大学 | A method of the image visual effect enhancing based on low-and high-frequency separation |
CN110232670B (en) * | 2019-06-19 | 2023-05-12 | 重庆大学 | Method for enhancing visual effect of image based on high-low frequency separation |
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 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106097274A (en) | A kind of adaptive fractional rank differential algorithm for image enhancement | |
Wang et al. | Dehazing for images with large sky region | |
CN108053417B (en) | lung segmentation device of 3D U-Net network based on mixed rough segmentation characteristics | |
CN106127688B (en) | A kind of super-resolution image reconstruction method and its system | |
CN103208097B (en) | Filtering method is worked in coordination with in the principal component analysis of the multi-direction morphosis grouping of image | |
CN105551010A (en) | Multi-focus image fusion method based on NSCT (Non-Subsampled Contourlet Transform) and depth information incentive PCNN (Pulse Coupled Neural Network) | |
CN103295204B (en) | A kind of image self-adapting enhancement method based on non-down sampling contourlet transform | |
CN110322404B (en) | Image enhancement method and system | |
CN104715461A (en) | Image noise reduction method | |
Yan et al. | Method to Enhance Degraded Image in Dust Environment. | |
CN102567973A (en) | Image denoising method based on improved shape self-adaptive window | |
CN104992421A (en) | OpenCL-based parallel optimization method of image de-noising algorithm | |
CN104021536A (en) | Self-adaptation SAR image and multispectral image fusion method | |
CN104616252B (en) | Digital image enhancement method based on NSCT and PCNN | |
CN104182939B (en) | Medical image detail enhancement method | |
CN108447033B (en) | Gray level image self-adaptive enhancement method based on singular value decomposition and longicorn whisker optimization algorithm | |
CN107516302A (en) | A kind of method of the mixed image enhancing based on OpenCV | |
CN103310414B (en) | Based on direction wave conversion and the image enchancing method of fuzzy theory | |
CN104616259B (en) | A kind of adaptive non-local mean image de-noising method of noise intensity | |
CN103985109A (en) | Feature-level medical image fusion method based on 3D (three dimension) shearlet transform | |
Lei et al. | Low-light image enhancement using the cell vibration model | |
CN104240208A (en) | Uncooled infrared focal plane detector image detail enhancement method | |
CN107169932A (en) | A kind of image recovery method based on Gauss Poisson mixed noise model suitable for neutron imaging system diagram picture | |
CN106780468A (en) | View-based access control model perceives the conspicuousness detection method of positive feedback | |
CN105279742A (en) | Quick image denoising method on the basis of partition noise energy estimation |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20161109 |