CN109523494A - A kind of multi-focus image fusing method - Google Patents
A kind of multi-focus image fusing method Download PDFInfo
- Publication number
- CN109523494A CN109523494A CN201811185806.7A CN201811185806A CN109523494A CN 109523494 A CN109523494 A CN 109523494A CN 201811185806 A CN201811185806 A CN 201811185806A CN 109523494 A CN109523494 A CN 109523494A
- Authority
- CN
- China
- Prior art keywords
- confidence map
- obtains
- corresponding position
- final
- initial
- 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
- 238000000034 method Methods 0.000 title claims abstract description 24
- 238000002156 mixing Methods 0.000 claims abstract description 8
- 238000001914 filtration Methods 0.000 claims abstract description 7
- 230000000694 effects Effects 0.000 abstract description 7
- 230000004927 fusion Effects 0.000 description 6
- 238000007500 overflow downdraw method Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000002955 isolation Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000000513 principal component analysis Methods 0.000 description 1
- 230000011514 reflex Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/20—Image enhancement or restoration using local operators
-
- 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
- G06T2207/20024—Filtering details
- G06T2207/20032—Median filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20212—Image combination
- G06T2207/20221—Image fusion; Image merging
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a kind of multi-focus image fusing methods, include the following steps, step 1 obtains initial confidence map P: finding out the spatial frequency of each pixel of source images A, B first, then compare the size of corresponding position spatial frequency, it is worth biggish corresponding position and sets 1, it is worth lesser corresponding position and sets 0, finally obtains 0-1 and scheme i.e. initial confidence map P;Step 2 refines initial confidence map P to obtain final confidence map T: carrying out median filtering refinement to initial confidence map P, then removing measure using wisp, further refinement obtains final confidence map T to it;Step 3 obtains blending image F: final blending image F is obtained by final confidence map T and source images A, B weighted sum.The present invention can be to not being registrated insensitive and can be effectively treated Boundary Effect problem between focusing and non-focusing.
Description
Technical field
The present invention relates to image fusion technology fields, and in particular to a kind of multi-focus image fusing method.
Background technique
In field of digital photography, the imaging device as digital single lens reflex camera is often difficult to shooting wherein institute
There is object to be all focused captured image.Typically, under certain optical lens focal length setting, only in the depth of field (DOF)
Object has sharp keen appearance in photo, and other objects are likely to obscure.Multi-focus image fusion is that one is fusions
A kind of popular technique of total focus image is obtained using the multiple images of the Same Scene of different focal length setting shooting.In recent years
Come, there has been proposed a variety of multi-focus image fusing methods, these methods are broadly divided into two major classes: transform domain method and space
Domain method.Multi-focus image fusing method based on multi-scale transform is most typical transform domain image fusion method, is mainly had
Based on fusion methods such as pyramid decomposition, wavelet transformation, multi-scale geometric analysis.Based on rarefaction representation and independent principal component analysis
Multi-focus image fusing method be different from the another kind of transform domain method based on multi-scale transform.Airspace method mainly has base
In block, it is based on image segmentation, multi-focus image fusing method pixel-based.
In the above-mentioned methods, transform domain multi-focus image fusing method key step are as follows: source images are converted first
Corresponding coefficient is obtained, then coefficient is merged using specific rule, finally obtains fusion coefficients progress inverse transformation
Merge multiple focussing image.Such methods assume that the conspicuousness information of original image can be extracted from these coefficients decomposed, and
And fusion coefficients are obtained based on the coefficients comparison of same position, therefore sensitivity is not registrated to image, it is easy to produce artificial effect;
Spatial domain multi-focus image fusing method generally first obtains confidence map, then obtains fusion figure using confidence map combination source images
Picture, it is insensitive to not being registrated, but it is typically due to cannot get accurate confidence map and focusing and non-focusing boundary effect is caused to ask
Topic.
Summary of the invention
Problem to be solved by this invention is: a kind of multi-focus image fusing method is provided, it can be insensitive to not being registrated
And Boundary Effect problem between focusing and non-focusing can be effectively treated.
The present invention in order to solve the above problem provided by technical solution are as follows: a kind of multi-focus image fusing method, feature
It is: includes the following steps,
Step 1, step 1 obtain initial confidence map P: finding out the spatial frequency of each pixel of source images A, B first, then
The size for comparing corresponding position spatial frequency is worth biggish corresponding position and sets 1, is worth lesser corresponding position and sets 0, finally obtains 0-
1 figure, that is, initial confidence map P;
Step 2 refines initial confidence map P to obtain final confidence map T: median filtering refinement is carried out to initial confidence map P,
Then removing measure using wisp, further refinement obtains final confidence map T to it;
Step 3 obtains blending image F: final blending image F is obtained by final confidence map T and source images A, B weighted sum
It arrives.
Compared with prior art, the invention has the advantages that the present invention can be insensitive to not being registrated, artificial effect will not be generated
It answers;Accurate confidence map can be obtained so as to which Boundary Effect problem between focusing and non-focusing is effectively treated.
Detailed description of the invention
The drawings described herein are used to provide a further understanding of the present invention, constitutes a part of the invention, this hair
Bright illustrative embodiments and their description are used to explain the present invention, and are not constituted improper limitations of the present invention.
Fig. 1 is the basic framework figure of the multi-focus image fusing method of the invention based on spatial frequency.
Specific embodiment
Carry out the embodiment that the present invention will be described in detail below in conjunction with accompanying drawings and embodiments, how the present invention is applied whereby
Technological means solves technical problem and reaches the realization process of technical effect to fully understand and implement.
Under specific embodiments of the present invention:
(1) initial confidence map P is obtained
The basic assumption that we obtain initial confidence map is: the spatial frequency values of pixel are higher, original image corresponding position
It is more clear.For convenience, source images A, B are respectively defined as A in the spatial frequency of the pixel of position (x, y)SF(x, y) and BSF(x,
y).It is H × L image I for a width size, the pixel value in position (x, y) pixel is M (x, y), spatial frequency is defined as:
In formula, line frequency RF and column frequency CF calculating are as follows:
Obtaining initial confidence map P, steps are as follows:
Step1: source images A, the A of B are calculated separatelySF(x, y) and BSF(x,y)。
Step2: the spatial frequency values for comparing corresponding position as the following formula obtain initial confidence map P.
Wherein, 1 representative is more clear than in source images 2 in source images 1 in this position pixel, and 0 represents in this
Position pixel is more clear than in source images 1 in source images 2.
(2) initial confidence map P is refined to obtain final confidence map T
After obtaining initial confidence map P, we remove measure using median filtering and wisp and refine to it.
Median filtering mainly removes in initial confidence map P that some isolation pixels and very small region, median filtering can
To be expressed as equation:
M (x, y)=Med { P (x-m, y-n), (m, n) ∈ w }
In formula, Med represents median filtering operation.
Wisp removal can delete small area object, can be expressed as following formula:
T=RSO { M, TH }
In formula, RSO represents wisp and removes operation, and TH be the number of pixels of setting, and what number of pixels was less than TH will be by
Removal.
In this patent, w is the sliding window of 8 × 8 sizes, and TH is adaptively set as source image pixels total numberIt adopts
Judge between element whether to be adjacent element with 8 neighborhoods.
(3) blending image F is obtained
After obtaining final confidence map T, we obtain blending image F using following formula:
F (x, y)=T (x, y) × A (x, y)+(1-T (x, y)) × B (x, y)
(1-T (x, y)) is the final confidence map of image B.
Only highly preferred embodiment of the present invention is described above, but is not to be construed as limiting the scope of the invention.This
Invention is not only limited to above embodiments, and specific structure is allowed to vary.All protection models in independent claims of the present invention
Interior made various change is enclosed to all fall in the scope of protection of the present invention.
Claims (1)
1. a kind of multi-focus image fusing method, it is characterised in that: include the following steps,
Step 1 obtains initial confidence map P: finding out the spatial frequency of each pixel of source images A, B first, then relatively more corresponding position
The size for setting spatial frequency is worth biggish corresponding position and sets 1, is worth lesser corresponding position and sets 0, and it is i.e. initial to finally obtain 0-1 figure
Confidence map P;
Step 2 refines initial confidence map P to obtain final confidence map T: carrying out median filtering refinement to initial confidence map P, then
Removing measure using wisp, further refinement obtains final confidence map T to it;
Step 3 obtains blending image F: final blending image F is obtained by final confidence map T and source images A, B weighted sum.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811185806.7A CN109523494A (en) | 2018-10-11 | 2018-10-11 | A kind of multi-focus image fusing method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811185806.7A CN109523494A (en) | 2018-10-11 | 2018-10-11 | A kind of multi-focus image fusing method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109523494A true CN109523494A (en) | 2019-03-26 |
Family
ID=65772006
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811185806.7A Pending CN109523494A (en) | 2018-10-11 | 2018-10-11 | A kind of multi-focus image fusing method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109523494A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111402183A (en) * | 2020-01-10 | 2020-07-10 | 北京理工大学 | Multi-focus image fusion method based on octave pyramid framework |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101976436A (en) * | 2010-10-14 | 2011-02-16 | 西北工业大学 | Pixel-level multi-focus image fusion method based on correction of differential image |
CN104077762A (en) * | 2014-06-26 | 2014-10-01 | 桂林电子科技大学 | Multi-focusing-image fusion method based on NSST and focusing area detecting |
CN104881855A (en) * | 2015-06-10 | 2015-09-02 | 北京航空航天大学 | Multi-focus image fusion method using morphology and free boundary condition active contour model |
CN108171676A (en) * | 2017-12-01 | 2018-06-15 | 西安电子科技大学 | Multi-focus image fusing method based on curvature filtering |
-
2018
- 2018-10-11 CN CN201811185806.7A patent/CN109523494A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101976436A (en) * | 2010-10-14 | 2011-02-16 | 西北工业大学 | Pixel-level multi-focus image fusion method based on correction of differential image |
CN104077762A (en) * | 2014-06-26 | 2014-10-01 | 桂林电子科技大学 | Multi-focusing-image fusion method based on NSST and focusing area detecting |
CN104881855A (en) * | 2015-06-10 | 2015-09-02 | 北京航空航天大学 | Multi-focus image fusion method using morphology and free boundary condition active contour model |
CN108171676A (en) * | 2017-12-01 | 2018-06-15 | 西安电子科技大学 | Multi-focus image fusing method based on curvature filtering |
Non-Patent Citations (1)
Title |
---|
QIAN JIANG 等: ""A novel multi-focus image fusion method based on stationary wavelet transform and local features of fuzzy sets"", 《IEEE》 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111402183A (en) * | 2020-01-10 | 2020-07-10 | 北京理工大学 | Multi-focus image fusion method based on octave pyramid framework |
CN111402183B (en) * | 2020-01-10 | 2023-08-11 | 北京理工大学 | Multi-focus image fusion method based on octave pyramid frame |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Zhou et al. | Spatio-temporal filter adaptive network for video deblurring | |
CN104616274B (en) | A kind of multi-focus image fusing method based on salient region extraction | |
CN104463817A (en) | Image processing method and device | |
DE102015102681B4 (en) | ALL-IN-FOCUS IMPLEMENTATION | |
CN114022823A (en) | Shielding-driven pedestrian re-identification method and system and storable medium | |
Huang et al. | Removing reflection from a single image with ghosting effect | |
CN112651911A (en) | High dynamic range imaging generation method based on polarization image | |
Hong et al. | Near-infrared image guided reflection removal | |
CN109523494A (en) | A kind of multi-focus image fusing method | |
Junginger et al. | Unpaired high-resolution and scalable style transfer using generative adversarial networks | |
CN115965844B (en) | Multi-focus image fusion method based on visual saliency priori knowledge | |
Tong et al. | Visual attention inspired distant view and close-up view classification | |
TWI805282B (en) | Methods and apparatuses of depth estimation from focus information | |
Guo et al. | Progressive Domain Translation Defogging network for real-world fog images | |
CN112508828A (en) | Multi-focus image fusion method based on sparse representation and guided filtering | |
CN112819742B (en) | Event field synthetic aperture imaging method based on convolutional neural network | |
Shihabudeen et al. | Euclidian norm based fusion strategy for multi focus images | |
Yang et al. | Video deblurring Via 3d CNN and fourier accumulation learning | |
CN106203244B (en) | A kind of determination method and device of lens type | |
US20220180544A1 (en) | Depth-of-field simulation via optical-based depth extraction | |
Zhao et al. | Image aesthetics enhancement using composition-based saliency detection | |
CN112200756A (en) | Intelligent bullet special effect short video generation method | |
Ataer-Cansizoglu et al. | Super-resolution of very low-resolution faces from videos | |
LIU et al. | A Lightweight and Efficient Infrared Pedestrian Semantic Segmentation Method | |
JP5928024B2 (en) | Image processing apparatus, image processing method, and program |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190326 |