CN101489033A - Image noise reduction method based on optimal original color signal - Google Patents
Image noise reduction method based on optimal original color signal Download PDFInfo
- Publication number
- CN101489033A CN101489033A CNA2008100023737A CN200810002373A CN101489033A CN 101489033 A CN101489033 A CN 101489033A CN A2008100023737 A CNA2008100023737 A CN A2008100023737A CN 200810002373 A CN200810002373 A CN 200810002373A CN 101489033 A CN101489033 A CN 101489033A
- Authority
- CN
- China
- Prior art keywords
- signal
- primary signal
- primary
- pass filter
- poorest
- 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.)
- Granted
Links
Images
Landscapes
- Processing Of Color Television Signals (AREA)
- Facsimile Image Signal Circuits (AREA)
- Picture Signal Circuits (AREA)
- Color Television Image Signal Generators (AREA)
- Color Image Communication Systems (AREA)
Abstract
The invention discloses a method for inhibiting an image noise based on an optimum primary color signal. A digital camera executes this noise inhibition method to inhibit the noise in the digital images mainly comprising: taking out the primary color signal in the digital image, taking out the high and low frequency signals of the optimum primary color signal through a high-pass filter and a low-pass filter, and the high and low frequency signals of the worst primary color signal, then mixing the high frequency signal of the optimum primary color signal and the worst primary color signal, and stacking the mixed signals to the low frequency signal of the worst primary color signal to obtain the digital image signal after the noise inhibition.
Description
Technical field
The invention discloses a kind of inhibition picture noise method, relate in particular to a kind of inhibition picture noise method based on best primary signal.
Background technology
The shooting quality of digital camera depends on photosensory assembly, as photosensitive coupling component (Charge CoupledDevice, CCD), complementarity matal-oxide semiconductor assembly (Complementary Metal-OxideSemiconductor, CMOS) photoperceptivity of assembly such as grade.Yet, when digital camera is taken with ISO pattern (HighISO), can produce picture noise unavoidably, these picture noises are the spot of random distribution normally.There is multiple digital camera to possess noise suppressed is arranged (Noise Reduction) function at present, makes digital camera when the ISO pattern, suppress the noise of photographic images.Existing image sensing assembly great majority have the inconsistent characteristic of each former chromatic sensitivity, or need do white balance because of the ambient light influence of color and handle, and take out the back each primary colors is imposed gain (gain) compensation, allow each primary colors reach desirable signal strength signal intensity at signal.But because of each primary colors gain compensation amount difference, the ratio that causes the noise of each primary signal to be exaggerated is also different.Big more when gaining, the noise of this primary colors is just big relatively more.In order to allow the approaching as far as possible primary colors preferably of relatively poor primary colors noise specific energy, past attempts proposes to remove at aberration field (domain) algorithm of noise.Its primary signal with the best is the center, produces the difference with other primary signal, difference is blured or computing such as neutralization again, restores difference at last.
But, with the noise of aberration field processing digital images, still must be excessive in the face of the characteristics of image processing, cause problems such as image detail can't keep.Especially, when noisiness is too big, keeps image detail and become the very work of difficulty.If will keep image detail, often need spend suitable operand.Single with regard to aberration field processing noise, its effect that can improve picture noise is still limited.
Summary of the invention
The object of the present invention is to provide a kind of inhibition picture noise method based on best primary signal, suppress the picture noise method by carrying out with digital camera, utilize the primary signal of optimum signal noise ratio, revise relatively poor primary signal, and then suppress the noise speckle of digital picture, solve and suppress the digital picture noise in the above-mentioned known technology and expend calculation resources and characteristics of image is handled the excessive problems such as details can't keep that cause.
To achieve these goals, the invention provides a kind of inhibition picture noise method, may further comprise the steps based on best primary signal:
At first, load digital picture, and take out a plurality of primary signals of digital picture.Afterwards, select the best primary signal in these primary signals, use high pass and low pass filter to be decomposed into high-frequency signal and low frequency signal; And select the poorest primary signal in these primary signals, and be decomposed into high-frequency signal and low frequency signal.Then, mix the best primary signal and the high-frequency signal of poor primary signal, and with mixed signal, be superimposed to the low frequency signal of poor primary signal, to obtain the data image signal after the noise suppressed.
According to the described inhibition picture noise method based on best primary signal of preferred embodiment of the present invention, wherein primary signal is primary red signal, green primary signal or blue primaries signal.
According to the described inhibition picture noise method based on best primary signal of preferred embodiment of the present invention, wherein low pass filter can be homogenizing (Average) filter or Gauss (Gaussian) filter; High pass filter can deduct the filter of the signal of original image after low-pass filtering for calculating the original image signal.
Another object of the present invention is to provide a kind of inhibition picture noise method based on best primary signal, this inhibition picture noise method is partly set weighted value by the low frequency of primary signal, allowing relatively poor primary signal obtain preferable noise ratio, and keep the thin portion details of primary signal.
For reaching above-mentioned inhibition noise purpose, the inhibition picture noise method based on best primary signal of the present invention may further comprise the steps:
At first, load digital picture, and take out a plurality of primary signals in the digital picture.Afterwards, select the best primary signal in these primary signals, use high pass and low pass filter to be decomposed into high-frequency signal and low frequency signal; And, select the poorest primary signal in these primary signals, and be decomposed into high-frequency signal and low frequency signal.Then, according to the best primary signal and the low frequency signal intensity of poor primary signal, setting signal hybrid weight ratio; According to this signal hybrid weight ratio, mix the best primary signal and the high-frequency signal of poor primary signal.And, mixed signal is superimposed to the low frequency signal of poor primary signal, to obtain the data image signal after the noise suppressed.
From the above mentioned, inhibition picture noise method based on best primary signal of the present invention, each primary signal in the Number of Decomposition word image is high-frequency signal and low frequency signal, utilization has the best primary signal of optimum signal noise ratio and revises other relatively poor primary signal, and with reference to each primary colors low frequency signal intensity, replace to fall the radio-frequency component of relatively poor primary signal,, have the not effect of fuzzy digit image simultaneously to reach the noise speckle that suppresses in the digital picture.
Describe the present invention below in conjunction with the drawings and specific embodiments, but not as a limitation of the invention.
Description of drawings
Fig. 1 is the inhibition picture noise method flow diagram based on best primary signal of a preferred embodiment of the present invention;
Fig. 2 is the oscillogram of the preferable primary signal of a preferred embodiment of the present invention;
Fig. 3 is the oscillogram of the relatively poor primary signal of a preferred embodiment of the present invention;
Fig. 4 is the oscillogram that the inhibition picture noise method of passing through best primary signal of a preferred embodiment of the present invention is produced;
Fig. 5 is the inhibition picture noise method flow diagram based on best primary signal of another preferred embodiment of the present invention.
Wherein, Reference numeral:
210 best primary signals
The low frequency part of 220 best primary signals
The HFS of 230 best primary signals
310 the poorest primary signals
320 low frequency part of poor primary signal
330 HFSs of poor primary signal
410 data image signals
Embodiment
First embodiment:
Fig. 1 is the inhibition picture noise method flow diagram based on best primary signal of a preferred embodiment of the present invention.Please refer to Fig. 1, present embodiment is carried out this inhibition picture noise method by digital camera, to suppress the noise in the digital picture.Described inhibition picture noise method based on best primary signal may further comprise the steps:
At first, load digital picture, and take out a plurality of primary signals (step S110) in the digital picture.Afterwards, select the best primary signal in these primary signals, and use high pass filter and low pass filter, best primary signal is decomposed into high-frequency signal and low frequency signal (step S120); Then, select the poorest primary signal in these primary signals, and use high pass filter and low pass filter, this poorest primary signal is decomposed into high-frequency signal and low frequency signal (step S130).At last, mix the best primary signal and the high-frequency signal of poor primary signal, and with mixed signal, be superimposed to the low frequency signal of poor primary signal, to obtain the data image signal (step S140) after the noise suppressed.
Wherein, described primary signal for example is primary red signal, green primary signal or blue primaries signal.In addition, aforesaid low pass filter is homogenizing (Average) filter or Gauss (Gaussian) filter; High pass filter is for can calculate the filter that the original image signal deducts the signal of original image after low-pass filtering.
Be the clear present embodiment of describing, now with the waveform explanation of primary signal.Fig. 2 is the oscillogram of the preferable primary signal of a preferred embodiment of the present invention.Please refer to Fig. 2, the initial data of digital picture for example represents with rgb format or yuv data form that it is the expression mode of spatial domain.After reading in tricolor signal, can get the oscillogram of the tricolor signal of a height vibration.Judge earlier what person is that relatively poor primary signal, what person are preferable primary signal in the tricolor signal.Judgment mode is: if the waveform tendency is comparatively vibrated, represent that then this primary signal is relatively poor primary signal; Otherwise, then be preferable primary signal.
In tricolor signal, take out best primary signal 210, and by low pass filter (Low PassFilter, LPF), for example homogenizing (Average) filter or Gauss (Gaussian) filter are decomposed into the low frequency part 220 of best primary signal by best primary signal 210; And by high pass filter (HighPass Filter HPF), for example for can calculate the filter that the original image signal deducts the signal of original image after low-pass filtering, is decomposed into best primary signal 210 HFS 230 of best primary signal.Similarly, after best primary signal 210 is decomposited low frequency part 220 and HFS 230, need again the poorest primary signal to be resolved.Fig. 3 is the oscillogram of the relatively poor primary signal of a preferred embodiment of the present invention.Please refer to Fig. 3, similarly, take out the low frequency part 320 of the poorest primary signal and the HFS 330 of poor primary signal from the poorest primary signal 310 by filter.
After the high and low frequency part of taking out respectively in the best, the poorest primary signal, promptly begin high frequency part with best primary signal, with reference to the low frequency signal intensity of each primary colors, replace the high frequency part of relatively poor primary signal.Fig. 4 is the oscillogram that the inhibition picture noise method of passing through best primary signal of a preferred embodiment of the present invention is produced.Please refer to Fig. 4, at first, the high frequency of best primary signal partly 230 is mixed (promptly overlapping according to the amplitude of wave form addition) with the high frequency part 330 of poor primary signal, obtain a transition waveform (not shown).After both waveforms are mixed into transition waveform (not shown), further with reference to the low frequency of the poorest primary signal partly 320.Afterwards, partly 320 be substrate according to this low frequency again, with transition waveform this low frequency part that superposes, and to adjust transition waveform after the stack be data image signal 410 after the noise suppressed.
Second embodiment:
Fig. 5 is the inhibition picture noise method flow diagram based on best primary signal of another preferred embodiment of the present invention.Please refer to Fig. 5, the execution mode of present embodiment is similar in appearance to above-mentioned first embodiment.But it is also setting signal hybrid weight ratio of present embodiment with the different part of first embodiment, adjusts the mixed proportion of waveform.In view of each primary signal (primary red signal, green primary signal or blue primaries signal) carries out gain compensation to primary signal in varing proportions, cause the substrate waveform different (meaning is that the waveform of low frequency signal is different) of the primary signal of gained.Therefore, when the low frequency with preferable primary signal replaces the high frequency part of relatively poor primary signal, go back low frequency signal intensity, setting signal hybrid weight ratio according to preferable primary signal and relatively poor primary signal.More clear so as to the digital picture that keeps the original signal details partly, make suppressing behind the noise.
Described inhibition picture noise method based on best primary signal may further comprise the steps: at first, load digital picture, and take out a plurality of primary signals (step S510) in the digital picture.Then, select the best primary signal in these primary signals, and use high pass filter and low pass filter, best primary signal is decomposed into high-frequency signal and low frequency signal (step S520); Afterwards, select the poorest primary signal in these primary signals, and use high pass filter and low pass filter, the poorest primary signal is decomposed into high-frequency signal and low frequency signal (step S530).Then, according to the best primary signal and the low frequency signal intensity of poor primary signal, the setting signal hybrid weight is than (step S540).At last, the basis signal hybrid weight is than mixing the best primary signal and the high-frequency signal of poor primary signal, and mixed signal is superimposed to the low frequency signal of poor primary signal, to obtain the data image signal (step S550) after the noise suppressed.
Certainly; the present invention also can have other various embodiments; under the situation that does not deviate from spirit of the present invention and essence thereof; those of ordinary skill in the art work as can make various corresponding changes and distortion according to the present invention, but these corresponding changes and distortion all should belong to the protection range of the appended claim of the present invention.
Claims (8)
1, a kind of inhibition picture noise method based on best primary signal is carried out this inhibition picture noise method with digital camera, to suppress the noise in the digital picture, it is characterized in that this inhibition picture noise method may further comprise the steps:
Load a digital picture, and take out several primary signals of this digital picture;
Select the best primary signal in this primary signal, and use a high pass filter and a low pass filter, should be decomposed into a high-frequency signal and a low frequency signal by the best primary signal;
Select the poorest primary signal in this primary signal, and use a high pass filter and a low pass filter, this poorest primary signal is decomposed into a high-frequency signal and a low frequency signal; And
Mix this best primary signal and this this high-frequency signal of the poorest primary signal, and, be superimposed to this this low frequency signal of the poorest primary signal, to obtain the data image signal after the noise suppressed mixed signal.
2, the inhibition picture noise method based on best primary signal according to claim 1 is characterized in that this primary signal is selected from the set of being made up of primary red signal, green primary signal and blue primaries signal.
3, the inhibition picture noise method based on best primary signal according to claim 1 is characterized in that this low pass filter is homogenizing filter or Gaussian filter.
4, the inhibition picture noise method based on best primary signal according to claim 1 is characterized in that this high pass filter is for can calculate the filter that the original image signal deducts the signal of original image after low-pass filtering.
5, a kind of inhibition picture noise method based on best primary signal is carried out this inhibition picture noise method with digital camera, to suppress the noise in the shot digital images, it is characterized in that this inhibition picture noise method may further comprise the steps:
Load a digital picture, and take out several primary signals of this digital picture;
Select the best primary signal in this primary signal, and use a high pass filter and a low pass filter, should be decomposed into a high-frequency signal and a low frequency signal by the best primary signal;
Select the poorest primary signal in this primary signal, and use a high pass filter and a low pass filter, this poorest primary signal is decomposed into a high-frequency signal and a low frequency signal;
According to this best primary signal and this this low frequency signal intensity of the poorest primary signal, set a signal hybrid weight ratio; And
Than mixing this high-frequency signal of the poorest primary signal of this best primary signal and this, and mixed signal is superimposed to this this low frequency signal of the poorest primary signal according to this signal hybrid weight, to obtain the data image signal after the noise suppressed.
6, the inhibition picture noise method based on best primary signal according to claim 5 is characterized in that this primary signal is selected from the set of being made up of primary red signal, green primary signal and blue primaries signal.
7, the inhibition picture noise method based on best primary signal according to claim 5 is characterized in that this low pass filter is homogenizing filter or Gaussian filter.
8, the inhibition picture noise method based on best primary signal according to claim 5 is characterized in that this high pass filter is for can calculate the filter that the original image signal deducts the signal of original image after low-pass filtering.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2008100023737A CN101489033B (en) | 2008-01-15 | 2008-01-15 | Image noise reduction method based on optimal original color signal |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2008100023737A CN101489033B (en) | 2008-01-15 | 2008-01-15 | Image noise reduction method based on optimal original color signal |
Publications (2)
Publication Number | Publication Date |
---|---|
CN101489033A true CN101489033A (en) | 2009-07-22 |
CN101489033B CN101489033B (en) | 2010-09-22 |
Family
ID=40891694
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN2008100023737A Expired - Fee Related CN101489033B (en) | 2008-01-15 | 2008-01-15 | Image noise reduction method based on optimal original color signal |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101489033B (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102754128A (en) * | 2009-11-25 | 2012-10-24 | 登塔尔图像科技公司 | Marker identification and processing in x-ray images |
US9082177B2 (en) | 2009-11-25 | 2015-07-14 | Dental Imaging Technologies Corporation | Method for tracking X-ray markers in serial CT projection images |
CN105390091A (en) * | 2015-11-04 | 2016-03-09 | 深圳市绿源半导体技术有限公司 | LED display enabling signal control system, LED display system and display driving method |
US9826942B2 (en) | 2009-11-25 | 2017-11-28 | Dental Imaging Technologies Corporation | Correcting and reconstructing x-ray images using patient motion vectors extracted from marker positions in x-ray images |
CN111770321A (en) * | 2019-03-27 | 2020-10-13 | 佳能株式会社 | Image processing apparatus, image capturing apparatus, image processing method, and storage medium |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9082182B2 (en) | 2009-11-25 | 2015-07-14 | Dental Imaging Technologies Corporation | Extracting patient motion vectors from marker positions in x-ray images |
US9082036B2 (en) | 2009-11-25 | 2015-07-14 | Dental Imaging Technologies Corporation | Method for accurate sub-pixel localization of markers on X-ray images |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0760566B2 (en) * | 1986-04-08 | 1995-06-28 | ソニー株式会社 | FM recording / reproducing system noise reduction circuit |
WO2001050420A1 (en) * | 1999-12-30 | 2001-07-12 | Applied Science Fiction, Inc. | Method and apparatus for eliminating colour crosstalk |
JP2008505523A (en) * | 2004-07-05 | 2008-02-21 | コーニンクレッカ フィリップス エレクトロニクス エヌ ヴィ | Camera color noise reduction method and circuit |
JP4470901B2 (en) * | 2006-03-31 | 2010-06-02 | ソニー株式会社 | Imaging apparatus, high-frequency component detection circuit, high-frequency component detection method, and computer program |
-
2008
- 2008-01-15 CN CN2008100023737A patent/CN101489033B/en not_active Expired - Fee Related
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102754128A (en) * | 2009-11-25 | 2012-10-24 | 登塔尔图像科技公司 | Marker identification and processing in x-ray images |
CN104598909A (en) * | 2009-11-25 | 2015-05-06 | 登塔尔图像科技公司 | Marker identification and processing in x-ray images |
US9082177B2 (en) | 2009-11-25 | 2015-07-14 | Dental Imaging Technologies Corporation | Method for tracking X-ray markers in serial CT projection images |
CN102754128B (en) * | 2009-11-25 | 2016-08-03 | 登塔尔图像科技公司 | Landmark identification in radioscopic image and process |
US9826942B2 (en) | 2009-11-25 | 2017-11-28 | Dental Imaging Technologies Corporation | Correcting and reconstructing x-ray images using patient motion vectors extracted from marker positions in x-ray images |
CN104598909B (en) * | 2009-11-25 | 2018-07-24 | 登塔尔图像科技公司 | Landmark identification in radioscopic image and processing |
CN105390091A (en) * | 2015-11-04 | 2016-03-09 | 深圳市绿源半导体技术有限公司 | LED display enabling signal control system, LED display system and display driving method |
CN105390091B (en) * | 2015-11-04 | 2017-12-08 | 深圳市绿源半导体技术有限公司 | LED shows enable signal control system, LED display system and display drive method |
CN111770321A (en) * | 2019-03-27 | 2020-10-13 | 佳能株式会社 | Image processing apparatus, image capturing apparatus, image processing method, and storage medium |
US11445158B2 (en) | 2019-03-27 | 2022-09-13 | Canon Kabushiki Kaisha | Image processing apparatus, image capturing apparatus, image processing method, and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN101489033B (en) | 2010-09-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN101489033B (en) | Image noise reduction method based on optimal original color signal | |
US8081239B2 (en) | Image processing apparatus and image processing method | |
US9165346B2 (en) | Method and apparatus for reducing image noise | |
CN204498282U (en) | There is the imaging system of visible ray photosensitive pixel and infrared light photosensitive pixel | |
US8675102B2 (en) | Real time denoising of video | |
US8723982B2 (en) | Image processing device, storage medium, and image processing method that corrects saturation, specific color and spatial frequency of an obtained image | |
EP2523160A1 (en) | Image processing device, image processing method, and program | |
CN105009168B (en) | Restore wave filter generating means and method, image processing apparatus and method, camera device | |
CN102045490B (en) | Image processing device and image processing method | |
CN105100636B (en) | A kind of image processing method and mobile terminal | |
US20090002530A1 (en) | Apparatus and method for processing images | |
CN101867826B (en) | Image processing apparatus and image processing method | |
CN104363434A (en) | Image processing apparatus | |
CN104471929B (en) | Color image sensor and camera head | |
US7801355B2 (en) | Image processing method, image processing device, semiconductor device, electronic apparatus, image processing program, and computer-readable storage medium | |
JP3540758B2 (en) | Horizontal contour signal generation circuit for single-chip color camera | |
US7898582B2 (en) | Image noise reduction method based on optimal primary-color signal | |
CN101646090A (en) | Luminance signal generation apparatus, luminance signal generation method, and image capturing apparatus | |
US7889279B2 (en) | Method and apparatus for suppressing cross-coloration in a video display device | |
WO2018100950A1 (en) | Image processing device, digital camera, image processing program, and recording medium | |
WO2007135798A1 (en) | Image processing device | |
CN106375316A (en) | Video image processing method and device | |
CN113840048A (en) | Method for intelligently adjusting brightness of four corners of camera picture | |
US20100303376A1 (en) | Circuit and method for processing image | |
US8345116B2 (en) | Noise reduction filter processing circuit, image processing circuit, imaging device, and storage medium storing noise reduction program |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20100922 Termination date: 20200115 |