CN100367763C - Method and device for clearing digital image noise - Google Patents
Method and device for clearing digital image noise Download PDFInfo
- Publication number
- CN100367763C CN100367763C CNB200510085206XA CN200510085206A CN100367763C CN 100367763 C CN100367763 C CN 100367763C CN B200510085206X A CNB200510085206X A CN B200510085206XA CN 200510085206 A CN200510085206 A CN 200510085206A CN 100367763 C CN100367763 C CN 100367763C
- Authority
- CN
- China
- Prior art keywords
- color
- values
- neighborhood pixels
- interested pixel
- value
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Processing Of Color Television Signals (AREA)
- Image Processing (AREA)
Abstract
This incvention relates to a method and a device for eliminating digntal image noises, which first of all selects a number of adjacent pixels around an interested pixel, then the color value of each adjacent pixel detracts that of the interested pixel, if the balance is higher than a preset critical value, then the color value of the adjacent pixel is replaced by that of the interested one, finally, the color values of all the adjacent pixels are averaged to get a mean value to replace that of the interested pixel.
Description
Technical field
The present invention particularly removes the method for noise about a kind of method and device for digitized video removing noise about a kind of high efficiency digitized video.
Background technology
Because the universal and people of computer make digital camera replace the status of traditional camera for the demand of convenience gradually.For digital camera, the image quality that is produced is an important evidence of judging the digital camera quality, so each tame manufacturer is devoted to the improvement of image quality invariably.In digital camera, if captured undressed image has noise, this noise will seriously influence the quality of image after treatment, this is because when the color-values of magnified image, also simultaneously the noise on it is amplified, therefore, before captured image is handled, need earlier the noise on it to be removed.
In known denoising sound method, Sigma's filtering (Sigma filter) method is preferable noise reset mode, Figure 1A, Figure 1B and Fig. 1 C are respectively 5 * 5 pel arrays 10,12 and 14 by Bayer (Bayer) colorful optical filter array selected green channel, red channel and blue channel, interested pixel in each array 10,12 and 14 centers is and will carries out the pixel that noise is removed, and this method is the color-values G of each neighborhood pixels of estimation earlier
1-G
12, R
1-R
8And B
1-B
8With the color-values G of interested pixel separately
C, R
CAnd B
CBetween the difference size whether greater than a default critical value TH; Then, with all and center pixel color values G
C, R
CAnd B
CAbsolute difference is equal to or less than the color-values of pixel of critical value TH and the color-values G of center pixel
CAdd overall average, to obtain the color-values that a mean value replaces the center pixel.
For more clearly demonstrating, with reference to Fig. 2 and Fig. 3 A, Fig. 3 B, Fig. 3 C and Fig. 3 D, wherein Fig. 2 is the flow chart of Sigma's filtering method, and Fig. 3 A, Fig. 3 B, Fig. 3 C and Fig. 3 D then are the example of being done according to Fig. 2 flow chart.Fig. 3 A shows 5 * 5 pel arrays of a green channel, in step 20, estimate the absolute value of color-values after deducting the color-values of interested pixel of all neighborhood pixels, in Fig. 3 A, center (3, the color-values of interested pixel 3) is 20, so the color-values of all pixels is deducted 20, the color-values of each pixel deducts absolute value after 20 shown in Fig. 3 B.In step 22, all absolute values are compared with the critical value TH that presets, suppose that TH is 10, step 24 is chosen the neighborhood pixels that is less than or equal to critical value TH with the color-values absolute difference of interested pixel then, and in this example, critical value TH is 10, so position (1,1), (1,5), (2,4), (3,5), (4,2) color-values that reaches the pixel of (4,4) is not considered, shown in Fig. 3 C.In step 26, the color-values of average selected neighborhood pixels and interested pixel and obtain a mean value 21, final step 28 replaces the color-values of interested pixels with mean value 21, shown in Fig. 3 D.In like manner red channel and blue channel also are the removings of reaching noise according to above-mentioned step.
Yet, by among Figure 1A, Figure 1B and Fig. 1 C as can be seen, when using Sigma's filtering method to carry out image processing, the number of pixels that is comprised in the pel array 10,12 and 14 of each color channel is different, comprise 12 neighborhood pixels in the green channel, and comprise 8 neighborhood pixels in redness and the blue channel, and all need count at every turn absolute value greater than and be equal to or less than the number of pixels of critical value TH, thereby the complexity when causing hardware to be carried out, moreover, judge that the number of pixels that is comprised in the array has also caused waste of time.
Therefore, a kind of is to be the institute Ji making things convenient for the method for hardware computing when the digitized video denoising sound.
Summary of the invention
One of purpose of the present invention is to provide a kind of method of removing noise for digitized video.
One of purpose of the present invention, other is to provide a kind of high efficiency digitized video denoising sound method.
According to the present invention, a kind ofly remove the method for noise for digitized video, at first around an interested pixel, choose the neighborhood pixels of a number; Secondly the color-values of each this neighborhood pixels is deducted the color-values of this interested pixel; Then with the color-values absolute difference of the color-values of neighborhood pixels and this interested pixel and a critical value relatively, if during greater than this critical value of presetting, the color-values of this neighborhood pixels is just with the color-values replacement of this interested pixel; Then the color-values with all neighborhood pixels on average obtains the color-values that a mean value replaces this interested pixel.
The present invention also points out a kind of device for digitized video removing noise, comprising: a buffer is for a temporary pel array; One first memory, the number of the neighborhood pixels that storage will be chosen; One second memory stores a critical value; And a processor, this processor comprises: pixel is chosen the unit, in order to the number that sets according to this first memory, by choosing interested pixel and neighborhood pixels thereof in this pel array; Absolute value calculation unit, and in order to the absolute value of the color-values difference of estimating each this neighborhood pixels and this interested pixel; Comparing unit, in order to relatively each this absolute value and this critical value, the color-values that replaces this neighborhood pixels with the color-values of this interested pixel during greater than the critical value that sets when the absolute value of the color-values difference of neighborhood pixels and this interested pixel is adjusted the color-values of this neighborhood pixels according to the presetting method adjustment; Average calculation unit obtains a mean value in order to the color-values of average these a plurality of neighborhood pixels, to replace the color-values of this interested pixel.
The present invention can be reached for digitized video and remove noise, and can high efficiency realization digitized video denoising sound.
Description of drawings
Figure 1A is 5 * 5 pel arrays by the selected green channel of Bayer colorful optical filter array;
Figure 1B is 5 * 5 pel arrays by the selected red channel of Bayer colorful optical filter array;
Fig. 1 C is 5 * 5 pel arrays by the selected blue channel of Bayer colorful optical filter array;
Fig. 2 is the flow chart of Sigma's filtering method;
Fig. 3 A shows the example of 5 * 5 pel arrays of a green channel;
The absolute value of the color-values difference of each neighborhood pixels and interested pixel among Fig. 3 B displayed map 3A;
The absolute value of the color-values difference of Fig. 3 C demonstration and interested pixel is less than or equal to the neighborhood pixels of default critical value;
Fig. 3 D shows the color-values of interested pixel after filter out noise;
Fig. 4 A is that the present invention is at 5 * 5 pel arrays of the selected green channel of a Bayer colorful optical filter array and selected neighborhood pixels;
Fig. 4 B is that the present invention is at 5 * 5 pel arrays of the selected red channel of a Bayer colorful optical filter array and selected neighborhood pixels;
Fig. 4 C is that the present invention is at 5 * 5 pel arrays of the selected blue channel of a Bayer colorful optical filter array and selected neighborhood pixels;
Fig. 5 shows the embodiment of denoising mixer of the present invention;
Fig. 6 is the flow chart of the method according to this invention;
Fig. 7 A shows the embodiment of 5 * 5 pel arrays of a green channel;
The absolute value of the color-values difference of each neighborhood pixels and interested pixel among Fig. 7 B displayed map 6A;
Fig. 7 C shows the schematic diagram after being replaced by the color-values of interested pixel greater than the neighborhood pixels color-values of default critical value with the absolute value of interested pixel color-values difference;
Fig. 7 D shows the color-values of interested pixel after filter out noise;
Fig. 8 is the curve chart of traditional noise sweep-out method gained;
Fig. 9 is the curve chart of Sigma's filtering method gained; And
Figure 10 is the curve chart of the inventive method gained.
Label declaration
5 * 5 pel arrays of 10 green channels
5 * 5 pel arrays of 12 red channels
5 * 5 pel arrays of 14 blue channels
The absolute value of the color-values difference of 20 each neighborhood pixels of estimation and interested pixel
22 with the absolute value of the color-values difference of each neighborhood pixels and interested pixel and a default critical value relatively
24 choose the neighborhood pixels that is less than or equal to critical value with the absolute value of the color-values difference of interested pixel
The color-values of 26 average selected neighborhood pixels and interested pixel obtains a mean value
28 color-values with mean value replacement interested pixel
5 * 5 pel arrays of 30 green channels
5 * 5 pel arrays of 32 red channels
5 * 5 pel arrays of 34 blue channels
40 remove the noise device
42 pel array buffers
44 operational stores
45 processors
46 memories
48 memories
50 choose neighborhood pixels around interested pixel
The absolute value of the color-values difference of 52 each neighborhood pixels of estimation and interested pixel
54 with the absolute value of the color-values of each neighborhood pixels and interested pixel and a default critical value relatively
56 will replace with the color-values of sense with interesting pixel with the absolute value of the color-values difference of the interested pixel color-values greater than the neighborhood pixels of critical value
The color-values of 58 average all neighborhood pixels obtains a mean value
59 color-values with mean value replacement interested pixel
The criteria for noise difference curve of 60 red color values when different grey scale average value
The criteria for noise difference curve of 62 blue color-values when different grey scale average value
The criteria for noise difference curve of 64 green tint values when different grey scale average value
The criteria for noise difference curve of 66 brightness when different grey scale average value
The criteria for noise difference curve of 70 red color values when different grey scale average value
The criteria for noise difference curve of 72 blue color-values when different grey scale average value
The criteria for noise difference curve of 74 green tint values when different grey scale average value
The criteria for noise difference curve of 76 brightness when different grey scale average value
The criteria for noise difference curve of 80 red color values when different grey scale average value
The criteria for noise difference curve of 82 blue color-values when different grey scale average value
The criteria for noise difference curve of 84 green tint values when different grey scale average value
The criteria for noise difference curve of 86 brightness when different grey scale average value
Embodiment
Fig. 4 A, Fig. 4 B and Fig. 4 C are respectively 5 * 5 pel arrays 30,32 and 34 of green channel, red channel and the blue channel chosen from the signal that a Bayer colorful optical filter array sensor is produced, the pixel in each array 30,32 and 34 centers is will carry out the interested pixel that noise is removed.Fig. 5 shows denoising mixer 40 of the present invention.The method according to this invention, the original color value I that is produced at Bayer colorful optical filter array sensor
ABehind the pel array buffer 42 of input unit 40, processor 45 is by original color value I
AIn choose the pel array G of green channel, red channel and blue channel respectively
i, R
iAnd B
i, for example the pel array shown in Fig. 4 A～Fig. 4 C 30,32 and 34 is stored in the operational store 44.To pel array G
i, R
iAnd B
iWhen carrying out the noise removing, earlier by the number Q that reads the neighborhood pixels that to choose in the memory 46, this numerical value is preestablished by input SET1, the preferably is 2 power power, wherein, when if the color of this interested pixel is green, selected neighborhood pixels will be arranged in a rhombus around this interested pixel, when if the color of this interested pixel is red or blue, selected neighborhood pixels will be arranged in a square around this interested pixel, shown in Fig. 4 A, Fig. 4 B and Fig. 4 C, in the present embodiment, Q is 8.Color-values G with neighborhood pixels in each passage
1-G
8, R
1-R
8And B
1-B
8, deduct separately sense and the color-values G of interesting pixel
C, R
CAnd B
CThen again by reading a critical value TH in the memory 48, this numerical value is preestablished by input SET2, with the absolute value of TH and all color-values difference relatively, when the color-values difference size of arbitrary neighborhood pixels during greater than this critical value TH, the color-values of this neighborhood pixels is with the color-values G of the center pixel of its place passage
C, R
COr B
CReplace, again neighborhood pixels all in each passage is added overall average respectively and obtain a mean value and replace the color-values G of the center pixel of passage separately
C, R
CAnd B
C, the G after last processor 45 will be handled
C, R
CAnd B
CPass back to pel array buffer 42, the color-values O behind the noise is removed in generation
A
For more clearly demonstrating above-mentioned method, with reference to the example of Fig. 6 and Fig. 7 A, Fig. 7 B, Fig. 7 C and Fig. 7 D, wherein Fig. 6 is the flow chart of the method according to this invention, and Fig. 7 A, Fig. 7 B, Fig. 7 C and Fig. 7 D be the example for being done according to Fig. 6 flow chart then.In step 50, around picture interested, choose 8 neighborhood pixels, the pixel of the green channel that is provided as Fig. 7 A, then step 52 is estimated the absolute value of the color-values difference of each neighborhood pixels and interested pixel, shown in Fig. 7 B, step 54 compares the absolute value and a critical value TH who presets of the color-values difference of each neighborhood pixels and interested pixel, in this embodiment, critical value TH is 10, next step 56 will replace greater than the color-values of the neighborhood pixels of critical value 10 color-values with interested pixel with the absolute value of the color-values difference of interested pixel, in Fig. 7 B, position (2,4), (3,5), (4,2) reach (4,4) neighborhood pixels and center (3, the absolute value of the color-values difference of sense interest pixel 3) is greater than 10, therefore the color-values of these pixels replaces with the color-values 20 of interested pixel, shown in Fig. 7 C, come the color-values of average all neighborhood pixels of step 58 again and obtain a mean value 22, final step 59 is with the color-values of these mean value 22 replacement interested pixels, shown in Fig. 7 D.
Because the number of selected neighborhood pixels is all identical in the present invention's color passage in office, therefore do not need to count again and deduct the number of pixels that absolute value behind this interested pixel is equal to or less than this critical value, so when doing multiplying, will make the easier processing of hardware.In addition, noise sweep-out method of the present invention can obtain the effect better than Sigma's filtering method of prior art.
Fig. 8, Fig. 9 and Figure 10 show the curve chart with traditional noise sweep-out method, Sigma's filtering method and method of the present invention simulation gained respectively, wherein X-axis is represented the mean value of all pixel gray levels, Y-axis is represented the standard deviation with normal color, and the standard deviation height of healing represents that then noise is more serious.In Fig. 8, curve 60 is the criteria for noise difference curve of red color value when different grey scale average value, curve 62 is the criteria for noise difference curve of blue color-values when different grey scale average value, curve 64 is the criteria for noise difference curve of green tint value when different grey scale average value, curve 66 is the criteria for noise difference curve of brightness when different grey scale average value, in the 9th figure, curve 70 is the criteria for noise difference curve of red color value when different grey scale average value, curve 72 is the criteria for noise difference curve of blue color-values when different grey scale average value, curve 74 is the criteria for noise difference curve of green tint value when different grey scale average value, curve 76 is the criteria for noise difference curve of brightness when different grey scale average value, in Figure 10, curve 80 is the criteria for noise difference curve of red color value when different grey scale average value, curve 82 is the criteria for noise difference curve of blue color-values when different grey scale average value, curve 84 is the criteria for noise difference curve of green tint value when different grey scale average value, and curve 86 is the criteria for noise difference curve of brightness when different grey scale average value.By the comparison of Fig. 8, Fig. 9 and Figure 10, the result of method gained of the present invention is better than traditional noise sweep-out method, and approximate with the result of Sigma filtering method gained, even better.
Table one is the comparing data of method of the present invention and traditional noise sweep-out method, wherein color change 1-13 by bright to dark change, Y represents brightness, R is the red color value, G is the green tint value, B is blue color-values, the result of the numeric representation method gained of the present invention in the four row fields of top and traditional noise sweep-out method gained result's difference, below four row fields are then represented the improvement degree of method of the present invention with respect to conventional method with percentage, value in the field is that negative value represents that the result of gained of the present invention is inferior to traditional noise sweep-out method, by table one as can be seen, have on the occasion of field far more than field with negative value, method of the present invention is much better than traditional noise sweep-out method.
Table one
| |
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | Mean value |
| The difference of Y | 0.21 | - 0.04 | 0 | 0.01 | 0.13 | 0.16 | 0.35 | 0.36 | 0.48 | 0.69 | 0.35 | 0.39 | 0.6 | 0.2838462 |
| The difference of R | 0.4 | - 0.06 | - 0.03 | 0.44 | 0.66 | 0.4 | 0.77 | 0.82 | 1.09 | 1.25 | 0.72 | 0.78 | 1.04 | 0.6369231 |
| The difference of G | 0.17 | - 0.02 | 0.02 | - 0.05 | 0.05 | 0.13 | 0.34 | 0.35 | 0.42 | 0.8 | 0.39 | 0.36 | 0.54 | 0.2692308 |
| The difference of B | 0.06 | 0.04 | 0.03 | 0.06 | 0.25 | 0.39 | 0.58 | 1 | 0.65 | 1.16 | 0.71 | 0.68 | 0.74 | 0.4884615 |
| The difference of |
22％ | -6％ | 0％ | 1％ | 12％ | 6％ | 27％ | 20％ | 24％ | 31％ | 18％ | 19％ | 37％ | 16％ |
| The difference of |
31％ | -6％ | -2 ％ | 24％ | 33％ | 13％ | 35％ | 28％ | 35％ | 33％ | 24％ | 27％ | 36％ | 24％ |
| The difference of |
20％ | -3％ | 2％ | -4 ％ | 5％ | 5％ | 23％ | 19％ | 20％ | 33％ | 20％ | 17％ | 32％ | 14％ |
| The difference of |
6％ | 4％ | 2％ | 4％ | 17％ | 12％ | 29％ | 37％ | 24％ | 36％ | 25％ | 25％ | 29％ | 19％ |
Table two is the comparing data of method of the present invention and Sigma's filtering method, similarly, color change 1-13 by bright to dark change, Y represents brightness, R is the red color value, G is the green tint value, B is blue color-values, the result of the numeric representation method gained of the present invention in the four row fields of top and the filtering method gained result's of Sigma difference, below four row fields are then represented the improvement degree of method of the present invention with respect to Sigma's filtering method with percentage, value in the field is that negative value represents that the result of gained of the present invention is inferior to Sigma's filtering method, by table two as can be seen, have on the occasion of field still more than field with negative value, method of the present invention slightly is better than Sigma's filtering method.
Table two
| Color change | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | Mean value |
| The difference of Y | 0.2 | - 0.02 | 0.03 | - 0.01 | - 0.01 | 0.08 | - 0.02 | 0.03 | 0.06 | 0.02 | 0.01 | 0.04 | 0 | 0.031538 |
| The difference of R | 0.23 | - 0.06 | 0 | 0.29 | 0.15 | 0.05 | - 0.01 | 0.08 | 0.24 | 0.19 | - 0.05 | 0.17 | 0.12 | 0.107692 |
| The difference of G | 0.2 | - 0.02 | 0.04 | - 0.04 | 0 | 0.06 | - 0.05 | 0.02 | - 0.02 | - 0.05 | 0 | - 0.05 | - 0.08 | 0.000769 |
| The difference of B | 0.04 | 0.03 | 0.1 | - 0.02 | - 0.01 | 0.02 | 0.05 | 0.14 | - 0.06 | - 0.07 | 0.02 | - 0.02 | 0.01 | 0.017692 |
| The difference of Y | 23％ | 3 ％ | 2％ | -1 ％ | -1 ％ | 3％ | -2 ％ | 2％ | 4％ | 1％ | 1％ | 2％ | 0％ | 2％ |
| The difference of R | 22％ | -6 ％ | 0％ | 17％ | 10％ | 2％ | -1 ％ | 4％ | 10％ | 7％ | -2 ％ | 7％ | 6％ | 6％ |
| The difference of G | 23％ | -3 ％ | 3％ | -3 ％ | 0％ | 2％ | -5 ％ | 1％ | -1 ％ | -3 ％ | 0％ | -3 ％ | -7％ | 0％ |
| The difference of B | 4％ | 3％ | 7％ | -1 ％ | -1 ％ | 1％ | 3％ | 8％ | -3 ％ | -3 ％ | 1％ | -1 ％ | 1％ | 1％ |
More than be stated as the purpose of illustrating for what preferred embodiment of the present invention was done, accurately be disclosed form and be not intended to limit the present invention, based on above instruction or to make an amendment or change from embodiments of the invention study be possible, embodiment explains orally principle of the present invention and allows to have the knack of this operator and utilize the present invention in practical application and select and narration with various embodiment, and technological thought attempt of the present invention is decided by following claim and equalization thereof.
Claims (6)
1. a method of removing digital image noise is characterized in that, comprises the following steps:
Around an interested pixel, choose the neighborhood pixels of some;
Estimate the absolute value of the color-values difference of each this neighborhood pixels and this interested pixel;
Relatively each this absolute value and a critical value are when the absolute value of the color-values difference of neighborhood pixels and this interested pixel replaces the color-values of this neighborhood pixels during greater than the critical value that sets with the color-values of this interested pixel; And
The color-values of average these a plurality of neighborhood pixels obtains a mean value, replaces the color-values of this interested pixel.
2. the method for claim 1 is characterized in that, this number is 2 power power.
3. the method for claim 1 is characterized in that, this neighborhood pixels of choosing is to be the rhombus configuration at center with this interested pixel.
4. the method for claim 1 is characterized in that, this neighborhood pixels of choosing is to be the square configuration at center with this interested pixel.
5. the method for claim 1 is characterized in that, this presetting method is to work as the absolute value of this difference greater than this critical value, replaces the color-values of this neighborhood pixels with the color-values of this interested pixel.
6. the device for digitized video removing noise is characterized in that, comprising:
One buffer is for a temporary pel array;
One first memory, the number of the neighborhood pixels that storage will be chosen;
One second memory stores a critical value; And
One processor, this processor comprises:
Pixel is chosen the unit, in order to the number that sets according to this first memory, by choosing interested pixel and neighborhood pixels thereof in this pel array;
Absolute value calculation unit is in order to the absolute value of the color-values difference of estimating each this neighborhood pixels and this interested pixel;
Comparing unit is in order to relatively each this absolute value and this critical value, when the absolute value of the color-values difference of neighborhood pixels and this interested pixel replaces the color-values of this neighborhood pixels during greater than the critical value that sets with the color-values of this interested pixel;
Average calculation unit obtains a mean value in order to the color-values of average these a plurality of neighborhood pixels, to replace the color-values of this interested pixel.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB200510085206XA CN100367763C (en) | 2005-07-19 | 2005-07-19 | Method and device for clearing digital image noise |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB200510085206XA CN100367763C (en) | 2005-07-19 | 2005-07-19 | Method and device for clearing digital image noise |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1901609A CN1901609A (en) | 2007-01-24 |
| CN100367763C true CN100367763C (en) | 2008-02-06 |
Family
ID=37657339
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB200510085206XA Expired - Fee Related CN100367763C (en) | 2005-07-19 | 2005-07-19 | Method and device for clearing digital image noise |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100367763C (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4470065A (en) * | 1982-03-25 | 1984-09-04 | Rca Corporation | Adaptive error concealment using horizontal information determination from adjacent lines |
| WO2001001675A2 (en) * | 1999-06-30 | 2001-01-04 | Logitech, Inc. | Video camera with major functions implemented in host software |
| CN1525402A (en) * | 2003-04-18 | 2004-09-01 | 北京中星微电子有限公司 | A dynamic detecting and compensating method for faulty pixel |
| CN1622637A (en) * | 2004-12-27 | 2005-06-01 | 北京中星微电子有限公司 | Image dead point and noise eliminating method |
-
2005
- 2005-07-19 CN CNB200510085206XA patent/CN100367763C/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4470065A (en) * | 1982-03-25 | 1984-09-04 | Rca Corporation | Adaptive error concealment using horizontal information determination from adjacent lines |
| WO2001001675A2 (en) * | 1999-06-30 | 2001-01-04 | Logitech, Inc. | Video camera with major functions implemented in host software |
| CN1525402A (en) * | 2003-04-18 | 2004-09-01 | 北京中星微电子有限公司 | A dynamic detecting and compensating method for faulty pixel |
| CN1622637A (en) * | 2004-12-27 | 2005-06-01 | 北京中星微电子有限公司 | Image dead point and noise eliminating method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1901609A (en) | 2007-01-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8295595B2 (en) | Generating full color images by demosaicing noise removed pixels from images | |
| CN101304484B (en) | Image forming apparatus, flaw pixel correction apparatus, processing method of the device | |
| US8089533B2 (en) | Fixed pattern noise removal circuit, fixed pattern noise removal method, program, and image pickup apparatus | |
| JP4290193B2 (en) | Image processing device | |
| JP2000236473A (en) | Image processing circuit for image input device | |
| US8164662B2 (en) | Image-processing device for color image data and method for the image processing of color image data | |
| CN103379343A (en) | Image processing device, image processing method, and program | |
| WO2006119302A2 (en) | System for scaling, yc adjustment, and color conversion in a display controller | |
| CN102655564A (en) | Image processing apparatus, image processing method, and program | |
| US20080291330A1 (en) | Advanced Noise Reduction in Digital Cameras | |
| US20150054991A1 (en) | Advanced noise reduction in digital cameras | |
| CN101534384A (en) | Image processing method, image processing device, and image capturing device | |
| CN107454285A (en) | Image noise estimation method and device and image extraction device | |
| WO2005112470A1 (en) | Image processing device and image processing program | |
| CN105763768B (en) | Image processing method, device and system | |
| CN101505368A (en) | Progressive-to-interlace conversion method, image processing apparatus, and imaging apparatus | |
| CN113632134B (en) | Method, computer readable storage medium, and HDR camera for generating high dynamic range image | |
| CN102291531A (en) | Image processing apparatus, image processing method, and program | |
| CN104410786A (en) | Image processing apparatus and control method for image processing apparatus | |
| EP2847998A1 (en) | Systems, methods, and computer program products for compound image demosaicing and warping | |
| CN100559834C (en) | Image processing apparatus, image processing method and program | |
| JP2004519142A (en) | Preventing green non-uniformity in image sensors | |
| US6281938B1 (en) | Signal processing system | |
| CN102801917A (en) | Image processing apparatus and method | |
| CN100367763C (en) | Method and device for clearing digital image 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 | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20080206 Termination date: 20170719 |