CN108694706A - A kind of video image denoising system - Google Patents
A kind of video image denoising system Download PDFInfo
- Publication number
- CN108694706A CN108694706A CN201810807237.9A CN201810807237A CN108694706A CN 108694706 A CN108694706 A CN 108694706A CN 201810807237 A CN201810807237 A CN 201810807237A CN 108694706 A CN108694706 A CN 108694706A
- Authority
- CN
- China
- Prior art keywords
- image
- noise
- denoising
- indicate
- video image
- 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.)
- Withdrawn
Links
- 238000011156 evaluation Methods 0.000 claims abstract description 21
- 230000000694 effects Effects 0.000 claims abstract description 11
- 239000011159 matrix material Substances 0.000 claims description 18
- 239000006002 Pepper Substances 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 4
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 238000003860 storage Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000004590 computer program Methods 0.000 description 2
- 230000007850 degeneration Effects 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- KLDZYURQCUYZBL-UHFFFAOYSA-N 2-[3-[(2-hydroxyphenyl)methylideneamino]propyliminomethyl]phenol Chemical compound OC1=CC=CC=C1C=NCCCN=CC1=CC=CC=C1O KLDZYURQCUYZBL-UHFFFAOYSA-N 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000012271 agricultural production Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 201000001098 delayed sleep phase syndrome Diseases 0.000 description 1
- 208000033921 delayed sleep phase type circadian rhythm sleep disease Diseases 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 238000000034 method Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G06T5/70—
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
Abstract
The present invention provides a kind of video image denoising systems, including model building module, noise remove module and denoising evaluation module, the model building module is for establishing noise of video image model, the noise remove module is for being removed noise of video image according to noise model, and the denoising evaluation module is for evaluating the denoising effect of noise remove module.Beneficial effects of the present invention are:Realize the accurate denoising of video image and the effective evaluation to denoising effect.
Description
Technical field
The present invention relates to Image Denoising Technology fields, and in particular to a kind of video image denoising system.
Background technology
Vision is one of the major way that the mankind obtain information.Image acts as abnormal important in human work and life
Role.Image procossing refers to the key technology that information is extracted from image, is had extensively in the various aspects such as industry and life
Application, such as agricultural production, oil exploration, biomedicine field.Image denoising is as the basic task in image procossing
Image analysis and understanding provide solid foundation.
Obtain in image, compression, the transmission stage, due to the influence of many factors such as environment, transmission channel, image can be by
The interference of noise, and image information is made to lose, generate distortion.The image of distortion is handled, image procossing will certainly be influenced
As a result, reduce the accuracy of extraction information, and then interfere the various judgements made accordingly and decision.Image denoising be will from containing
Noise jamming is removed in noisy picture signal, to recover image actual signal, and then ensures further image procossing
With precision of analysis.Existing denoising model filters out research greatly both for white Gaussian noise for impulsive noise
Seldom.
Invention content
In view of the above-mentioned problems, the present invention is intended to provide a kind of video image denoising system.
The purpose of the present invention is realized using following technical scheme:
A kind of video image denoising system is provided, including mould is evaluated in model building module, noise remove module and denoising
Block, the model building module are used for for establishing noise of video image model, the noise remove module according to noise model
Noise of video image is removed, the denoising evaluation module is for evaluating the denoising effect of noise remove module;
The model building module is for establishing noise of video image model, specially:
Noise of video image model is expressed as by each frame image in video sequence as an image block:
H=F+N+M
In formula, F indicates the clean image array of not Noise, F={ f1,f2,…,fl},fiIndicate i-th of clean image
Block, l indicate that number of image frames, i=1,2 ..., l, N indicate impulsive noise matrix, N={ n1,n2,…,nl, niIndicate fiIt is corresponding
Impulsive noise, i=1,2 ..., l, M indicate Gaussian noise matrix, M={ m1,m2,…,ml, miIndicate fiCorresponding Gaussian noise,
I=1,2 ..., l, H are noise-containing image array, H={ h1,h2,…,hl, hiIndicate fiCorresponding noise-containing figure
Picture, i=1,2 ..., l.
Beneficial effects of the present invention are:The accurate denoising of video image and the effective evaluation to denoising effect are realized, is adopted
It uses each image that inclines as an image block, and there is in video between consecutive frame internal structure similitude, each other structure
At similar image block, modeling in this way does not have to the size for determining image block not only, it is thus also avoided that the measurement of similitude, to contribute to
Reduce the interference of noise.
Description of the drawings
Using attached drawing, the invention will be further described, but the embodiment in attached drawing does not constitute any limit to the present invention
System, for those of ordinary skill in the art, without creative efforts, can also obtain according to the following drawings
Other attached drawings.
Fig. 1 is the structural schematic diagram of the present invention;
Reference numeral:
Model building module 1, noise remove module 2, denoising evaluation module 3.
Specific implementation mode
The invention will be further described with the following Examples.
Referring to Fig. 1, a kind of video image denoising system of the present embodiment, including model building module 1, noise remove module
2 and denoising evaluation module 3, the model building module 1 is for establishing noise of video image model, the noise remove module 2
For being removed to noise of video image according to noise model, the denoising evaluation module 3 is used for noise remove module 2
Denoising effect is evaluated;
The model building module 1 is for establishing noise of video image model, specially:
Noise of video image model is expressed as by each frame image in video sequence as an image block:
H=F+N+M
In formula, F indicates the clean image array of not Noise, F={ f1,f2,…,fl},fiIndicate i-th of clean image
Block, l indicate that number of image frames, i=1,2 ..., l, N indicate impulsive noise matrix, N={ n1,n2,…,nl, niIndicate fiIt is corresponding
Impulsive noise, i=1,2 ..., l, M indicate Gaussian noise matrix, M={ m1,m2,…,ml, miIndicate fiCorresponding Gaussian noise,
I=1,2 ..., l, H are noise-containing image array, H={ h1,h2,…,hl, hiIndicate fiCorresponding noise-containing figure
Picture, i=1,2 ..., l.
What the algorithm of existing removal impulsive noise was carried out both for single image, and video is the orderly sequence of several images
Row, the time redundancy that vision signal itself has are ignored, when this results in the impulsive noise in existing algorithm removal video
Inefficiency, the present embodiment algorithm use each image that inclines as an image block, and have between consecutive frame in video interior
Portion's structural similarity constitutes similar image block each other, and modeling in this way does not have to the size for determining image block not only, it is thus also avoided that
The measurement of similitude, to help to reduce the interference of noise.
Preferably, the noise remove module 2 includes the first denoising submodule and the second denoising submodule, and described first goes
Submodule of making an uproar is used for for removing video image Gaussian noise, the second denoising submodule to the video after removal Gaussian noise
Image is handled, and video image impulsive noise is removed;
The first denoising submodule is for removing video image Gaussian noise, specially:Gauss is removed to every frame image
Noise obtains the video image after removal Gaussian noise:
S=F+N
In formula, S is the only image array containing impulsive noise, S={ s1,s2,…,sl, siIndicate fiIt is corresponding only to contain
The image of impulsive noise, i=1,2 ..., l;
The second denoising submodule is used to handle the video image after removal Gaussian noise, removes video image
Impulsive noise, specially:
For the video sequence containing l frames, l groups are classified as, to each frame image, centered on the image, front and back each n
The similar image block is ranked sequentially composition matrix, minimizes the matrix by similar image block of the frame image as the image
Order obtains a handling result of the image, since 2n+1 result can simply add by every frame image by processing 2n+1 times
Weight average removes the result after impulsive noise as the image;
It seeks after all frame image removal impulsive noises as a result, obtaining the video image G, Z=after removal impulsive noise
{z1,z2,…,zl},ziIndicate that the image after i-th of removal impulsive noise, l indicate number of image frames, i=1,2 ..., l.
Due to the time of video towering remaining property, structure is similar between adjacent image, if muting video image is formed
One matrix, then the matrix has low-rank.And the matrix that the video image with noise is constituted is exactly the low-rank matrix part
The matrix of the contaminated degeneration of element, removal video noise seek to recover low-rank matrix from the matrix of degeneration.This is preferably
Embodiment is somebody's turn to do every frame image by carrying out simple weighted average to minimizing rank of matrix, and to handling result
Image removes the video image after impulsive noise, handles all frame images, has obtained the video after removal impulsive noise
Image.
Preferably, the denoising evaluation module 3 is for evaluating the denoising effect of noise remove module, specially:
Evaluation points are defined using following formula:
In formula, Q indicates evaluation points,The image z after removal salt-pepper noise is indicated respectivelyiNot Noise
Clean image fiAverage brightness,The image z after removal salt-pepper noise is indicated respectivelyiThe not clean figure of Noise
As fiBrightness variance,Indicate the image z after removal salt-pepper noiseiThe not clean image f of NoiseiLuminance standard
Difference, PSNR indicate the image z after removal salt-pepper noiseiThe not clean image f of NoiseiY-PSNR;The evaluation because
Son is bigger, indicates that the denoising effect of described image denoising module is better.
Subjective assessment mode is that the visual effect of perception algorithm handling result is gone with human eye, and from subjectivity, measure algorithm is up
It is no to reach expected, if noise remove is clean, if to cause fuzzy etc..Subjective assessment and estimator itself are in close relations,
The possible otherness of evaluation conclusion that different observers obtains is very big.This preferred embodiment imitates denoising by defining evaluation points
Fruit is evaluated, and has fully considered the Y-PSNR of denoising image and the structural similarity of denoising image and original image,
The accurate evaluation for realizing noise remove effect overcomes the different caused evaluation differences of estimator in subjective assessment.
Through the above description of the embodiments, those skilled in the art can be understood that it should be appreciated that can
To realize the embodiments described herein with hardware, software, firmware, middleware, code or its any appropriate combination.For hardware
It realizes, processor can be realized in one or more the following units:Application-specific integrated circuit (ASIC), digital signal processor
(DSP), digital signal processing appts (DSPD), programmable logic device (PLD), field programmable gate array (FPGA), processing
Device, controller, microcontroller, microprocessor, other electronic units or combinations thereof designed for realizing functions described herein.
For software implementations, some or all of embodiment flow can instruct relevant hardware to complete by computer program.
When realization, above procedure can be stored in computer-readable medium or as the one or more on computer-readable medium
Instruction or code are transmitted.Computer-readable medium includes computer storage media and communication media, wherein communication media packet
It includes convenient for transmitting any medium of computer program from a place to another place.Storage medium can be that computer can
Any usable medium of access.Computer-readable medium can include but is not limited to RAM, ROM, EEPROM, CD-ROM or other
Optical disc storage, magnetic disk storage medium or other magnetic storage apparatus or can be used in carry or store with instruction or data
The desired program code of structure type simultaneously can be by any other medium of computer access.
Finally it should be noted that the above embodiments are merely illustrative of the technical solutions of the present invention, rather than the present invention is protected
The limitation of range is protected, although being explained in detail to the present invention with reference to preferred embodiment, those skilled in the art answer
Work as understanding, technical scheme of the present invention can be modified or replaced equivalently, without departing from the reality of technical solution of the present invention
Matter and range.
Claims (5)
1. a kind of video image denoising system, which is characterized in that evaluated including model building module, noise remove module and denoising
Module, the model building module are used for for establishing noise of video image model, the noise remove module according to noise mode
Type is removed noise of video image, and the denoising evaluation module is for commenting the denoising effect of noise remove module
Valence;
The model building module is for establishing noise of video image model, specially:
Noise of video image model is expressed as by each frame image in video sequence as an image block:
H=F+N+M
In formula, F indicates the clean image array of not Noise, F={ f1, f2..., fl, fiIndicate i-th of clean image block, l tables
Show that number of image frames, i=1,2 ..., l, N indicate impulsive noise matrix, N={ n1, n2..., nl, niIndicate fiCorresponding pulse is made an uproar
Sound, i=1,2 ..., l, M indicate Gaussian noise matrix, M={ m1, m2..., ml, miIndicate fiCorresponding Gaussian noise, i=1,
2 ..., l, H are noise-containing image array, H={ h1, h2..., hl, hiIndicate fiCorresponding noise-containing image, i=
1,2 ..., l.
2. video image denoising system according to claim 1, which is characterized in that the noise remove module includes first
Denoising submodule and the second denoising submodule, the first denoising submodule is for removing video image Gaussian noise, and described the
Two denoising submodules are used to handle the video image after removal Gaussian noise, remove video image impulsive noise.
3. video image denoising system according to claim 2, which is characterized in that the first denoising submodule is for going
Except video image Gaussian noise, specially:Gaussian noise is removed to every frame image, obtains the video figure after removal Gaussian noise
Picture:
S=F+N
In formula, S is the only image array containing impulsive noise, S={ s1, s2..., sl, siIndicate fiIt is corresponding only to contain pulse
The image of noise, i=1,2 ..., l.
4. video image denoising system according to claim 3, which is characterized in that the second denoising submodule for pair
Video image after removal Gaussian noise is handled, and removes video image impulsive noise, specially:
For the video sequence containing l frames, l groups are classified as, to each frame image, centered on the image, front and back each n frames figure
As the similar image block as the image, the similar image block is ranked sequentially composition matrix, the rank of matrix is minimized, obtains
To a handling result of the image, since 2n+1 result can be carried out simple weighted and put down by every frame image by processing 2n+1 times
The result after impulsive noise is removed as the image;
It seeks after all frame image removal impulsive noises as a result, obtaining video image G, the Z={ z after removal impulsive noise1,
z2..., zl, ziIndicate that the image after i-th of removal impulsive noise, l indicate number of image frames, i=1,2 ..., l.
5. video image denoising system according to claim 4, which is characterized in that the denoising evaluation module is used for making an uproar
The denoising effect of sound removal module is evaluated, specially:
Evaluation points are defined using following formula:
In formula, Q indicates evaluation points,The image z after removal salt-pepper noise is indicated respectivelyiThe not clean figure of Noise
As fiAverage brightness,The image z after removal salt-pepper noise is indicated respectivelyiThe not clean image f of Noisei's
Brightness variance,Indicate the image z after removal salt-pepper noiseiThe not clean image f of NoiseiLuminance standard it is poor, PSNR
Indicate the image z after removal salt-pepper noiseiThe not clean image f of NoiseiY-PSNR;The evaluation points are bigger,
Indicate that the denoising effect of described image denoising module is better.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810807237.9A CN108694706A (en) | 2018-07-21 | 2018-07-21 | A kind of video image denoising system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810807237.9A CN108694706A (en) | 2018-07-21 | 2018-07-21 | A kind of video image denoising system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108694706A true CN108694706A (en) | 2018-10-23 |
Family
ID=63851759
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810807237.9A Withdrawn CN108694706A (en) | 2018-07-21 | 2018-07-21 | A kind of video image denoising system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108694706A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112488959A (en) * | 2020-12-16 | 2021-03-12 | 杭州电子科技大学 | CMOS image sensor signal denoising method based on dynamic stochastic resonance |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106846268A (en) * | 2017-01-04 | 2017-06-13 | 温州大学 | A kind of Gaussian pulse mixed image noise remove method |
| CN107123094A (en) * | 2017-03-23 | 2017-09-01 | 四川精目科技有限公司 | A kind of mixing Poisson, the video denoising method of gaussian sum impulsive noise |
| CN107895503A (en) * | 2017-11-07 | 2018-04-10 | 钟永松 | A kind of unattended parking farm monitoring system |
| US20180182072A1 (en) * | 2016-12-22 | 2018-06-28 | Ateme | Hybride denoising of images and videos based on interest metrics |
-
2018
- 2018-07-21 CN CN201810807237.9A patent/CN108694706A/en not_active Withdrawn
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180182072A1 (en) * | 2016-12-22 | 2018-06-28 | Ateme | Hybride denoising of images and videos based on interest metrics |
| CN106846268A (en) * | 2017-01-04 | 2017-06-13 | 温州大学 | A kind of Gaussian pulse mixed image noise remove method |
| CN107123094A (en) * | 2017-03-23 | 2017-09-01 | 四川精目科技有限公司 | A kind of mixing Poisson, the video denoising method of gaussian sum impulsive noise |
| CN107895503A (en) * | 2017-11-07 | 2018-04-10 | 钟永松 | A kind of unattended parking farm monitoring system |
Non-Patent Citations (2)
| Title |
|---|
| 张智勇: "应用残差总变分及低秩表示的视频去噪算法", 《信号处理》 * |
| 朱艳丽: "基于低秩分解和形态成分分析的WMSN视频去噪算法研究", 《中国优秀硕士论文全文数据库 信息科技辑》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112488959A (en) * | 2020-12-16 | 2021-03-12 | 杭州电子科技大学 | CMOS image sensor signal denoising method based on dynamic stochastic resonance |
| CN112488959B (en) * | 2020-12-16 | 2024-02-09 | 杭州电子科技大学 | CMOS image sensor signal denoising method based on dynamic stochastic resonance |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wang et al. | Local phase coherence and the perception of blur | |
| CN103093419B (en) | A kind of method of detected image sharpness and device | |
| CN107085833B (en) | Remote sensing images filtering method based on the equal intermediate value fusion of gradient inverse self-adaptive switch | |
| CN103530896A (en) | Image compression and detail enhancement method for infrared image | |
| CN104685538A (en) | Systems and methods for reducing noise in video streams | |
| Liu et al. | Contrast enhancement using non-overlapped sub-blocks and local histogram projection | |
| CN105809633B (en) | Remove the method and device of color noise | |
| Tekalp et al. | Quantitative analysis of artifacts in linear space-invariant image restoration | |
| CN111654719B (en) | Video micro-motion detection method based on permutation entropy algorithm | |
| CN108694706A (en) | A kind of video image denoising system | |
| CN110349101A (en) | A kind of multiple dimensioned united Wavelet image denoising method | |
| CN106991670B (en) | no-reference noise image quality evaluation method and system | |
| Li et al. | Bionic vision-based synthetic aperture radar image edge detection method in non-subsampled contourlet transform domain | |
| Smolka | Robust sharpening vector median filter | |
| Biswas | Impulse Noise Detection and Removal Method Based on Modified Weighted Median | |
| CN108848352A (en) | A kind of cloud service video monitoring system | |
| Sheta | Restoration of medical images using genetic algorithms | |
| CN108769586A (en) | A kind of accurate monitoring system of monitoring | |
| CN108955699A (en) | A kind of onboard navigation system | |
| Coumar et al. | Image restoration using filters and image quality assessment using reduced reference metrics | |
| CN108765348A (en) | A kind of pilotless automobile | |
| Li | A class of multi-shell min/max median filters | |
| Gu et al. | SATD based fast intra prediction for HEVC | |
| CN111291830B (en) | Method for improving glass surface defect detection efficiency and accuracy | |
| Meher | Development of some novel nonlinear and adaptive digital image filters for efficient noise suppression |
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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20181023 |