CN106447621A - Video image denoising method and apparatus based on fuzzy connection principle - Google Patents
Video image denoising method and apparatus based on fuzzy connection principle Download PDFInfo
- Publication number
- CN106447621A CN106447621A CN201610750378.2A CN201610750378A CN106447621A CN 106447621 A CN106447621 A CN 106447621A CN 201610750378 A CN201610750378 A CN 201610750378A CN 106447621 A CN106447621 A CN 106447621A
- Authority
- CN
- China
- Prior art keywords
- fuzzy
- video image
- fuzzy connection
- denoising
- complement
- 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 23
- 230000000295 complement effect Effects 0.000 claims description 55
- 239000000284 extract Substances 0.000 claims description 7
- 239000000203 mixture Substances 0.000 claims description 5
- 230000000694 effects Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 5
- 230000008859 change Effects 0.000 description 3
- 230000007935 neutral effect Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 239000006002 Pepper Substances 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000006866 deterioration Effects 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000004899 motility Effects 0.000 description 1
- 238000013486 operation strategy Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000011946 reduction process Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G06T5/70—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformation in the plane of the image
- G06T3/40—Scaling the whole image or part thereof
- G06T3/4038—Scaling the whole image or part thereof for image mosaicing, i.e. plane images composed of plane sub-images
Abstract
The invention discloses a video image denoising method and apparatus based on a fuzzy connection principle. The method includes acquiring a video image signal and extracting the first frame of image of the video image signal; according to the preset fuzzy connection topology graph and the first frame of image, calculating the connection parameters of the fuzzy connection topology graph, wherein the fuzzy connection topology graph includes a first subsystem, a second subsystem, a plurality of amplifiers, and the fuzzy connection relations among the first subsystem, the second subsystem, the plurality of amplifiers, and the fuzzy connection relations include serial connection, parallel connection, and feedback; and according to the connection parameters and the fuzzy connection topology graph, denoising the video image signal and obtaining the denoised video image. The video image denoising method and apparatus can filter the high-frequency signal noise, and is high in adaptability and robustness.
Description
Technical field
The present invention relates to field of computer technology, more particularly to a kind of video image denoising side based on fuzzy connection principle
Method and device.
Background technology
Video image is the most frequently used information carrier in today's society, but in the acquisition in image, transmission or storage process
Usually by the interference of various noises and impact, image deterioration can be made.In order to obtain more preferable video image, need to image
Noise reduction process is carried out, which needs to keep raw information integrity as far as possible, removes useless information in signal again, to improve
Picture quality, increases signal to noise ratio.
Existing video image noise-removed technology mainly obtains preferable effect in low-dimensional signal and image processing, does not but apply to
In higher-dimension signal and image processing.As sub-band processing technology or fuzzy image processing technology, sub-band processing technology is by image and video
Denoising is carried out after being divided into several sub-bands, and fuzzy image processing technology can only process low frequency signal.These image procossing
Method declines can the detailed information of image when image is changed, and Real time signal process is poor, and computation complexity is high time-consuming
Long.
Content of the invention
The embodiment of the present invention proposes a kind of video image denoising method based on fuzzy connection principle and device, can realize height
The filtering of frequency signal noise, adaptability and strong robustness.
The embodiment of the present invention provides a kind of video image denoising method based on fuzzy connection principle, including:
Video signal is obtained, extracts the first two field picture of the video signal;
Complement and first two field picture are opened up according to default fuzzy connection, is calculated the acquisition fuzzy connection and open up complement
Connecting quantity;Wherein, the fuzzy connection open up complement by the first subsystem, the second subsystem, several amplifiers and three it
Between fuzzy connection relation composition;The fuzzy connection relation includes:Series, parallel and feedback;
Complement being opened up according to the Connecting quantity and the fuzzy connection, denoising is carried out to the video signal, obtains
Video image after denoising.
Further, first subsystem is Wiener filter;
Second subsystem is median filter.
Further, described complement and first two field picture are opened up according to default fuzzy connection, calculate and obtain the mould
The Connecting quantity of complement is opened up in paste connection, specially:
The fuzzy connection relation in complement is opened up according to the fuzzy connection, using first two field picture as training set, is counted
Calculate the Connecting quantity for obtaining several amplifiers.
Further, described complement is opened up according to the Connecting quantity and the fuzzy connection, to the video signal
Denoising is carried out, the video image after denoising is obtained, specially:
The Connecting quantity is substituted into the fuzzy connection complement is opened up, and using first subsystem and second son
System, carries out denoising to the follow-up two field picture of the video signal, obtains the video image after denoising.
Correspondingly, the embodiment of the present invention also provides a kind of video image denoising device based on fuzzy connection principle, including:
Video image acquisition module, Connecting quantity computing module and video image denoising module;
Wherein, the video image acquisition module is used for obtaining video signal, extracts the video signal
First two field picture;
The Connecting quantity computing module is used for opening up complement and first two field picture according to default fuzzy connection, calculates
Obtain the Connecting quantity that the fuzzy connection opens up complement;Wherein, the fuzzy connection opens up complement by the first subsystem, the second subsystem
System, the fuzzy connection relation composition between several amplifier and threes;The fuzzy connection relation includes:Series, parallel and
Feedback;
The video image denoising module opens up complement according to the Connecting quantity and the fuzzy connection, to the video figure
As signal carries out denoising, the video image after denoising is obtained.
Further, first subsystem is Wiener filter;
Second subsystem is median filter.
Further, the Connecting quantity computing module specifically for opening up the fuzzy company in complement according to the fuzzy connection
Relation is connect, using first two field picture as training set, calculates the Connecting quantity for obtaining several amplifiers.
Further, the video image denoising module is opened up specifically for the Connecting quantity is substituted into the fuzzy connection
Complement, and using first subsystem and second subsystem, the follow-up two field picture of the video signal is gone
Make an uproar, obtain the video image after denoising.
Implement the embodiment of the present invention, have the advantages that:
A kind of video image denoising method based on fuzzy connection principle provided in an embodiment of the present invention and device, first obtain
Video signal, and the first two field picture of complement and video signal is opened up by the advance fuzzy connection for building, calculate phase
The Connecting quantity that answers, finally opens up complement using the Connecting quantity and fuzzy connection, carries out denoising to video signal, and acquisition is gone
Video image after making an uproar.The sub-band processing technology for using compared to prior art or fuzzy image processing technology can only process low frequency
Signal, the fuzzy connection of technical scheme opens up complement for neutral net shape structure, can effectively filter out high-frequency signal, use
Scope is wide.In addition, the fuzzy connection relation of the present invention includes series, parallel and feedback so that the system has very strong suitable
Ying Xing, improves the robustness of whole system.
Description of the drawings
Fig. 1 is a kind of flow process of embodiment of the video image denoising method based on fuzzy connection principle that the present invention is provided
Schematic diagram;
Fig. 2 is that a kind of fuzzy connection of embodiment that the present invention is provided opens up complement;
Fig. 3 is that a kind of order of connection of embodiment of the coefficient of connection that the fuzzy connection that the present invention is provided opens up complement is illustrated
Figure;
Fig. 4 is the denoising effect comparison diagram that the present invention is provided;
Fig. 5 is a kind of structure of embodiment of the video image denoising device based on fuzzy connection principle that the present invention is provided
Schematic diagram.
Specific embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation is described, it is clear that described embodiment is only a part of embodiment of the present invention, rather than whole embodiments.It is based on
Embodiment in the present invention, it is every other that those of ordinary skill in the art are obtained under the premise of creative work is not made
Embodiment, belongs to the scope of protection of the invention.
Referring to Fig. 1, it is a kind of embodiment of the video image denoising method based on fuzzy connection principle that the present invention is provided
Schematic flow sheet.As shown in figure 1, the method comprising the steps of 101 to 103, specific as follows:
Step 101:Video signal is obtained, extracts the first two field picture of video signal.
In the present embodiment, before filtering and noise reduction, need first to obtain video signal, video signal is by some frames
Image constitutes.The present invention extracts the first frame of video signal as training set, to improve the effect of denoising.
Step 102:Complement and the first two field picture are opened up according to default fuzzy connection, is calculated acquisition fuzzy connection and open up complement
Connecting quantity;Wherein, fuzzy connection opens up complement by the first subsystem, the second subsystem, between several amplifier and threes
Fuzzy connection relation constitutes;Fuzzy connection relation includes:Series, parallel and feedback.
In the present embodiment, the fuzzy connection for being, referring to Fig. 2 and Fig. 3, Fig. 2, a kind of embodiment that the present invention is provided opens up benefit
Figure.Fig. 3 is a kind of order of connection schematic diagram of embodiment of the coefficient of connection that the fuzzy connection that the present invention is provided opens up complement.As schemed
Shown in 2, the fuzzy connection opens up complement by the first subsystem, the second subsystem, the fuzzy company between several amplifier and threes
Connect relation composition.As one kind citing of the present embodiment, the first subsystem TiFor Wiener filter;Second subsystem TjFor intermediate value
Wave filter.A={ x1, y1, x2, y2 }. a total of four elements in set A, the form shown in Fig. 3 be according to x1, y1, x2,
The connected mode (series, parallel, feedback) of y2 draws the order of connection of 16 parameters.
In the present embodiment, if TiBy series connection and TjConnection, be in other words, zj=Tj(Ti(ui)) and uj=0, then
Corresponding fuzzy payoff is all zero except A (yi,xj)=1;If TjBy series connection and TiConnection, then corresponding fuzzy category
Property value is all zero except A (yj,xi)=1.
In the present embodiment, if TiIn parallel and TjConnection, be in other words, zi=Ti(ui),zj=Tj(ui) and uj=
0, then corresponding fuzzy payoff is all zero except A (xi,xj)=1.
In the present embodiment, if TiBy feedback and TjConnection, be in other words, zi=Ti(ui+Tj(zi)) and uj=0, that
Corresponding fuzzy payoff is all zero except A (yi,xj)=1 and A (yj,xi)=1.
In the present embodiment, step 102 is specially:The fuzzy connection relation in complement is opened up according to fuzzy connection, by first
Two field picture calculates, as training set, the Connecting quantity for obtaining several amplifiers.
Step 103:Complement being opened up according to Connecting quantity and fuzzy connection, denoising is carried out to video signal, obtains denoising
Video image afterwards.
In the present embodiment, step 103 is specially:Connecting quantity substitution fuzzy connection is opened up complement, and using the first son
System and the second subsystem, carry out denoising to the follow-up two field picture of video signal, obtain the video image after denoising.
In order to better illustrate operation principle and the effect of the present invention, it is the denoising effect that the present invention is provided referring to Fig. 4, Fig. 4
Fruit comparison diagram.As shown in figure 4, the video signal is the greyscale video sequence " Claire " of one group of 8 100 frame, which includes
White Gaussian noise and impulsive noise, the size of each frame is 288*360.Video sequence is filtered comprising 3*3 wiener by one
The fuzzy system of ripple device and 3*3 median filter is filtered.The present invention opens up complement using fuzzy connection, by the of video sequence
One frame draws the coefficient of connection of 16 amplifiers as training set, then by the coefficient of connection that draws to after in video sequence
Continuous frame carries out noise reduction.Output quality is assessed finally by Y-PSNR:
From the simulation result of Fig. 4, the denoising effect of the present invention is than traditional filial generation treatment technology or traditional fuzzy connection denoising
Effect will be got well.
The first subsystem in the present embodiment and the second subsystem can be changed according to practical situation, illustrated as described above
In Gaussian noise when becoming salt-pepper noise, need to only change the type of the first subsystem, will Wiener filter be changed to another kind
Wave filter, the first frame of re -training draws new Connecting quantity, need not change the attachment structure that fuzzy connection opens up complement, user
Just.
Only need in the present invention train the first frame to obtain 16 amplifiers (parameter), and by this 16 parameters according to fuzzy
Link topology, connection Wiener filter and median filter, and without ambiguity solution module, video image is changed, fortune
Wider with scope, high-frequency signal noise can also be effectively filtered out.As fuzzy connection topological diagram is the structure of neutral net shape, if
Meter is with more motility.
Further, since Gaussian noise and impulsive noise are changed with the change of channel, fuzzy connection is opened up in complement
Feedback link so that the system has very strong adaptability, and new amplifier coefficient can be by old amplifier coefficient iteration
Self adaptation is produced, with good robustness and real-time.
Accordingly, referring to Fig. 5, Fig. 5 be the present invention provide the video image denoising device based on fuzzy connection principle
A kind of structural representation of embodiment.As shown in figure 5, the device includes:Video image acquisition module 501, Connecting quantity is calculated
Module 502 and video image denoising module 503.
Wherein, video image acquisition module 501 is used for obtaining video signal, extracts the first frame of video signal
Image.
Connecting quantity computing module 502 is used for opening up complement and the first two field picture according to default fuzzy connection, calculates and obtains
Fuzzy connection opens up the Connecting quantity of complement;Wherein, fuzzy connection open up complement by the first subsystem, the second subsystem, several put
Big fuzzy connection relation composition between device and three;Fuzzy connection relation includes:Series, parallel and feedback.
Video image denoising module 503 opens up complement according to Connecting quantity and fuzzy connection, and video signal is gone
Make an uproar, obtain the video image after denoising.
In the present embodiment, the first subsystem is Wiener filter;Second subsystem is median filter.
In the present embodiment, Connecting quantity computing module 502 specifically for opening up the fuzzy company in complement according to fuzzy connection
Relation is connect, using the first two field picture as training set, calculates the Connecting quantity for obtaining several amplifiers.
In the present embodiment, video image denoising module 503 opens up complement specifically for Connecting quantity is substituted into fuzzy connection,
And using the first subsystem and second subsystem, denoising is carried out to the follow-up two field picture of video signal, after obtaining denoising
Video image.
In sum, a kind of video image denoising method based on fuzzy connection principle provided in an embodiment of the present invention and dress
Putting, video signal is first obtained, and the first frame figure of complement and video signal is opened up by the advance fuzzy connection for building
Picture, calculates corresponding Connecting quantity, finally opens up complement using the Connecting quantity and fuzzy connection, video signal is gone
Make an uproar, obtain the video image after denoising.The sub-band processing technology for using compared to prior art or fuzzy image processing technology are only
Low frequency signal can be processed, the fuzzy connection of technical scheme is opened up complement for neutral net shape structure, can effectively filter out height
Frequency signal, operation strategies are wide.In addition, the fuzzy connection relation of the present invention includes series, parallel and feedback so that the system has
There is very strong adaptability, improve the robustness of whole system.
The above is the preferred embodiment of the present invention, it is noted that for those skilled in the art
For, under the premise without departing from the principles of the invention, some improvements and modifications can also be made, these improvements and modifications are also considered as
Protection scope of the present invention.
Claims (8)
1. a kind of video image denoising method based on fuzzy connection principle, it is characterised in that include:
Video signal is obtained, extracts the first two field picture of the video signal;
Complement and first two field picture are opened up according to default fuzzy connection, calculates the connection for obtaining that the fuzzy connection opens up complement
Parameter;Wherein, the fuzzy connection opens up complement by the first subsystem, the second subsystem, between several amplifier and threes
Fuzzy connection relation constitutes;The fuzzy connection relation includes:Series, parallel and feedback;
Complement being opened up according to the Connecting quantity and the fuzzy connection, denoising is carried out to the video signal, obtains denoising
Video image afterwards.
2. the video image denoising method based on fuzzy connection principle according to claim 1, it is characterised in that described
One subsystem is Wiener filter;
Second subsystem is median filter.
3. the video image denoising method based on fuzzy connection principle according to claim 2, it is characterised in that described
Complement and first two field picture are opened up according to default fuzzy connection, the Connecting quantity for obtaining that the fuzzy connection opens up complement are calculated,
Specially:
The fuzzy connection relation in complement is opened up according to the fuzzy connection, using first two field picture as training set, calculating is obtained
Obtain the Connecting quantity of several amplifiers.
4. the video image denoising method based on fuzzy connection principle according to claim 3, it is characterised in that described
Complement being opened up according to the Connecting quantity and the fuzzy connection, denoising is carried out to the video signal, obtains regarding after denoising
Frequency image, specially:
The Connecting quantity is substituted into the fuzzy connection complement is opened up, and using first subsystem and second subsystem
System, carries out denoising to the follow-up two field picture of the video signal, obtains the video image after denoising.
5. a kind of video image denoising device based on fuzzy connection principle, it is characterised in that include:Video image obtains mould
Block, Connecting quantity computing module and video image denoising module;
Wherein, the video image acquisition module is used for obtaining video signal, extracts the first of the video signal
Two field picture;
The Connecting quantity computing module is used for opening up complement and first two field picture according to default fuzzy connection, calculates and obtains
The fuzzy connection opens up the Connecting quantity of complement;Wherein, the fuzzy connection open up complement by the first subsystem, the second subsystem,
Fuzzy connection relation composition between several amplifier and threes;The fuzzy connection relation includes:Series, parallel and anti-
Feedback;
The video image denoising module opens up complement according to the Connecting quantity and the fuzzy connection, and the video image is believed
Denoising number is carried out, obtains the video image after denoising.
6. the video image denoising device based on fuzzy connection principle according to claim 5, it is characterised in that described
One subsystem is Wiener filter;
Second subsystem is median filter.
7. the video image denoising device based on fuzzy connection principle according to claim 6, it is characterised in that the company
Parameter calculating module is connect specifically for opening up the fuzzy connection relation in complement according to the fuzzy connection, by first two field picture
As training set, the Connecting quantity for obtaining several amplifiers is calculated.
8. the video image denoising device based on fuzzy connection principle according to claim 7, it is characterised in that described regard
Frequency image denoising module opens up complement specifically for the Connecting quantity is substituted into the fuzzy connection, and uses first subsystem
System and second subsystem, carry out denoising to the follow-up two field picture of the video signal, obtain the video image after denoising.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610750378.2A CN106447621A (en) | 2016-08-29 | 2016-08-29 | Video image denoising method and apparatus based on fuzzy connection principle |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610750378.2A CN106447621A (en) | 2016-08-29 | 2016-08-29 | Video image denoising method and apparatus based on fuzzy connection principle |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106447621A true CN106447621A (en) | 2017-02-22 |
Family
ID=58182730
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610750378.2A Pending CN106447621A (en) | 2016-08-29 | 2016-08-29 | Video image denoising method and apparatus based on fuzzy connection principle |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106447621A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11006108B2 (en) | 2017-09-14 | 2021-05-11 | Samsung Electronics Co., Ltd. | Image processing apparatus, method for processing image and computer-readable recording medium |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101013503A (en) * | 2007-01-26 | 2007-08-08 | 清华大学 | Method for segmenting abdominal organ in medical image |
US20150363634A1 (en) * | 2014-06-17 | 2015-12-17 | Beijing Kuangshi Technology Co.,Ltd. | Face Hallucination Using Convolutional Neural Networks |
-
2016
- 2016-08-29 CN CN201610750378.2A patent/CN106447621A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101013503A (en) * | 2007-01-26 | 2007-08-08 | 清华大学 | Method for segmenting abdominal organ in medical image |
US20150363634A1 (en) * | 2014-06-17 | 2015-12-17 | Beijing Kuangshi Technology Co.,Ltd. | Face Hallucination Using Convolutional Neural Networks |
Non-Patent Citations (1)
Title |
---|
WING-KUEN LING 等: "Video Denoising Using Fuzzy-connectedness Principles", 《PROCEEDINGS OF 2001 LNTERNATIONAL SYMPOSIUM ON INTELLIGENT MULTIMEDIA, VIDEO AND SPEECH PROCESSING》 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11006108B2 (en) | 2017-09-14 | 2021-05-11 | Samsung Electronics Co., Ltd. | Image processing apparatus, method for processing image and computer-readable recording medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105825484B (en) | A kind of depth image denoising and Enhancement Method based on deep learning | |
CN101980284B (en) | Two-scale sparse representation-based color image noise reduction method | |
CN111915513B (en) | Image denoising method based on improved adaptive neural network | |
CN107645621A (en) | A kind of method and apparatus of Video processing | |
CN105006001B (en) | A kind of method for evaluating quality for having ginseng image based on nonlinear organization similarity deviation | |
CN103077506A (en) | Local and non-local combined self-adaption image denoising method | |
CN105243650A (en) | Method and system for removing stripe noise on the basis of Wavelet transform and Fourier transform | |
CN107993200A (en) | Picture noise level estimation method based on deep learning | |
CN103208097A (en) | Principal component analysis collaborative filtering method for image multi-direction morphological structure grouping | |
CN111260591A (en) | Image self-adaptive denoising method based on attention mechanism | |
CN106960417A (en) | A kind of noise based on the notable structure of image obscures Image Blind mesh Deconvolution Method | |
CN109345609B (en) | Method for denoising mural image and generating line drawing based on convolutional neural network | |
CN101719268A (en) | Generalized Gaussian model graph denoising method based on improved Directionlet region | |
CN107451981A (en) | Picture noise level estimation method based on DCT and gradient covariance matrix | |
CN111723239A (en) | Multi-mode-based video annotation method | |
CN108877831A (en) | Blind source separating fast method and system based on multi-standard fusion frequency point screening | |
CN113723171A (en) | Electroencephalogram signal denoising method based on residual error generation countermeasure network | |
CN110010144A (en) | Voice signals enhancement method and device | |
CN106447621A (en) | Video image denoising method and apparatus based on fuzzy connection principle | |
CN104732498A (en) | Thresholding image denoising method based on nonsubsampled Contourlet transformation | |
CN107133938A (en) | Robust image fusion method based on wavelet coefficient statistical model | |
CN108876736A (en) | A kind of image alias removing method based on FPGA | |
CN107798667A (en) | Face Enhancement Method based on residual error study | |
CN102497576A (en) | Full-reference image quality assessment method based on mutual information of Gabor features (MIGF) | |
CN105872315B (en) | A kind of video denoising method for mixed noise |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170222 |