CN101686389B - Image down sampling method and system - Google Patents
Image down sampling method and system Download PDFInfo
- Publication number
- CN101686389B CN101686389B CN 200810168487 CN200810168487A CN101686389B CN 101686389 B CN101686389 B CN 101686389B CN 200810168487 CN200810168487 CN 200810168487 CN 200810168487 A CN200810168487 A CN 200810168487A CN 101686389 B CN101686389 B CN 101686389B
- Authority
- CN
- China
- Prior art keywords
- component
- input picture
- wavelet decomposition
- down sampling
- image down
- 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
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
The invention discloses image down sampling method and system, wherein the image down sampling method comprises the following steps: determining an image dissection layer level according to a predetermined image down sampling rate; carrying out a wavelet dissection on each component of an input image according to the determined image dissection layer level; determining a normalizing factor used for the normalization of lowest-frequency wavelet coefficients of all components after the wavelet dissection of the input image according to the lowest-frequency wavelet coefficient of each component after the wave dissection; normalizing the lowest-frequency wavelet coefficients of all components after the wavelet dissection to a time domain according to the determined normalized factor. The lowest-frequency wavelet coefficient of each component after the wavelet decomposition can be transformed to the time domain directly through the invention, thereby reducing the computation burden of the image down sampling method and system, improving transformation speed and saving computation resources.
Description
Technical field
The present invention relates to image processing field, relate more specifically to a kind of image down sampling method and system.
Background technology
Image down sampling is that a kind of image resolution ratio that reduces is to show, to store image and/or the technology of transmission etc.
The traditional image Downsapling method has directly adopted the interpolation method such as arest neighbors interpolation or bilinear interpolation in time domain; The Downsapling method of this time domain can cause the overlapping of pixel color; Thereby produce so-called mole effect, influence the visual quality of down-sampled images.And, when sample rate is higher instantly, according to above-mentioned interpolation method, the discontinuous neighbor of can sampling out, thus make and have rough scene in the big or small adjusted image.
Because good local characteristic at time domain and frequency domain; Wavelet transformation has kept the most information of received image signal in its low frequency coefficients piece; Thereby after one deck decomposes, formed size be merely received image signal half the, with the similar signal of received image signal.In addition, because low-frequency signals is the smooth of primary signal,, aspect visual quality, be superior to the interpolation method in the time domain so the image that utilizes the relative superiority or inferiority sample rate to sample out has level and smooth scene.
Summary of the invention
One or more problems in view of the above the invention provides a kind of new image down sampling method and system.
A kind of image down sampling method according to the embodiment of the invention may further comprise the steps: confirm the picture breakdown level according to predetermined image down sampling rate; According to determined picture breakdown level each component of input picture is carried out wavelet decomposition; According to the low-limit frequency wavelet coefficient of each component after the wavelet decomposition, confirm to be used for to after the wavelet decomposition of input picture important low-limit frequency wavelet coefficient carry out normalized normalization factor; And according to determined normalization factor after with the wavelet decomposition of input picture important low-limit frequency wavelet coefficient normalize to time domain.
Wherein, confirm picture breakdown level: L=log through following formula
2(M), wherein, M presentation video down-sampling rate, the L presentation video decomposes level, log
2Expression is the logarithm at the end with 2.Adopt biorthogonal Daubechies9/7 tap filter that each component of input picture is carried out wavelet decomposition.Can carry out the down-sampling of level and vertical both direction to input picture, and be used for image down sampling rate that each component to input picture carries out the horizontal direction down-sampling and be used for the image down sampling rate that each component to input picture carries out the vertical direction down-sampling and equate.Further, can carry out the wavelet decomposition of horizontal direction earlier to each component of input picture, again the horizontal direction wavelet decomposition result of each component of input picture carried out the wavelet decomposition of vertical direction.In addition, with after the wavelet decomposition of input picture the maximum value of important low-limit frequency wavelet coefficient as normalization factor.
A kind of image down sampling system according to the embodiment of the invention comprises: decompose level and confirm the unit, be used for confirming the picture breakdown level according to predetermined image down sampling rate; The wavelet decomposition performance element is used for according to determined picture breakdown level each component of input picture being carried out wavelet decomposition; Normalization factor is confirmed the unit, is used for the low-limit frequency wavelet coefficient according to each component after the wavelet decomposition, confirm to be used for to after the wavelet decomposition of input picture important low-limit frequency wavelet coefficient carry out normalized normalization factor; And component normalization performance element, be used for according to determined normalization factor after with the wavelet decomposition of input picture important low-limit frequency wavelet coefficient normalize to time domain.
Wherein, decompose level and confirm that the unit confirms picture breakdown level: L=log through following formula
2(M), wherein, in the down-sampling rate of level and vertical direction, the L presentation video decomposes level, log to the M presentation video respectively
2Expression is the logarithm at the end with 2.The wavelet decomposition performance element adopts biorthogonal Daubechies9/7 tap filter to come each component of input picture is carried out wavelet decomposition.In addition; This image down sampling system can carry out the down-sampling of level and vertical both direction to input picture, and is used for image down sampling rate that each component to input picture carries out the horizontal direction down-sampling and is used for the image down sampling rate that each component to input picture carries out the vertical direction down-sampling and equates.Further, the wavelet decomposition performance element can carry out the wavelet decomposition of horizontal direction earlier to each component of input picture, the horizontal direction wavelet decomposition result of each component of input picture is carried out the wavelet decomposition of vertical direction again.In addition, normalization factor confirm after the unit is with the wavelet decomposition of input picture the maximum value of important low-limit frequency wavelet coefficient confirm as normalization factor.
Because the low-frequency signals that obtains through wavelet decomposition has kept the most information of input picture, so the image that draws through the present invention is the additive method that is superior to aspect the visual quality directly carrying out in time domain (especially for than high sampling rate).In addition, the present invention can convert directly to time domain with the low-limit frequency wavelet coefficient of each component after the wavelet decomposition, so can save the operand of image down sampling method and system, improves conversion rate, saves computational resource.
Description of drawings
With reference to following description, can understand above-mentioned and other advantages of the present invention better in conjunction with the drawings, wherein:
Fig. 1 shows the flow chart according to the image down sampling method of the embodiment of the invention;
Fig. 2 shows the sketch map of the broad sense wavelet decomposition when picture breakdown level L equals 2;
Fig. 3 shows the sketch map of the processing of LazyRow function execution;
Fig. 4 shows the sketch map of the processing of FirstLiftRow function execution; And
Fig. 5 shows the block diagram according to the image down sampling system of the embodiment of the invention.
Embodiment
Below in conjunction with accompanying drawing, describe embodiment of the present invention in detail.
Fig. 1 shows the flow chart according to the image down sampling method of the embodiment of the invention.As shown in Figure 1, this image down sampling method is divided into following steps: S102, confirm the picture breakdown level according to predetermined image down sampling rate; S104 carries out wavelet decomposition according to determined picture breakdown level to each component of input picture; S106, according to the low-limit frequency wavelet coefficient of each component after the wavelet decomposition, confirm to be used for to after the wavelet decomposition of input picture important low-limit frequency wavelet coefficient carry out normalized normalization factor; And S108, according to determined normalization factor after with the wavelet decomposition of input picture important low-limit frequency wavelet coefficient normalize to time domain.Respectively above step S102 to S108 is elaborated below:
Step S102 confirms picture breakdown level L.Suppose, will utilize down-sampling rate (that is above-described image down sampling rate) M received image signal to be carried out down-sampling in level and vertical direction.For example, M is 2 power, and can confirm L:L=log through following formula
2(M), wherein, log
2Expression is the logarithmic function at the end with 2.
Step S104, the picture breakdown level L that respectively each component wavelet decomposition of input picture is confirmed in the step S102.Fig. 2 has provided the sketch map of the broad sense wavelet decomposition when L equals 2.Wherein, adopted biorthogonal Daubechies9/7 tap filter to decompose each component of input picture.In each wavelet decomposition level, at first input picture is carried out the line translation of horizontal direction, then the horizontal direction wavelet decomposition result of input picture is carried out the rank transformation of vertical direction.After the decomposition of having accomplished a wavelet decomposition level, upper left 1/4th that decomposite is the low frequency piece of current wavelet decomposition level.And if also do not reach given in advance wavelet decomposition level, then the low frequency piece of this current wavelet decomposition level can also be as the input picture of next wavelet decomposition level.For first wavelet decomposition level, input picture is initial picture content.Suppose that input picture component C is Row
COK * Col
CThe piece of row, the picture breakdown level is L, then decomposable process is: at first picture content C is carried out the wavelet transformation of first level, comprising line translation to C, and to line translation result's rank transformation.After the wavelet transformation of first level was accomplished, this moment, 1/4th in the upper left corner of C was exactly the wavelet low frequency rate coefficient piece of first level, corresponding Row with correspondence
CAnd Col
CReduce by half, decompose level and successively decrease.After aforesaid operations is accomplished, if predetermined decomposition layer progression does not reach (L>0), then with the low frequency piece of current layer as input signal, further carry out the wavelet decomposition of next level.For example, can realize through following block the decomposable process of input picture component:
WaveletDecompose(C，RowC，ColC，L)
{
While(L>0)
{
DecomposeRow(C，RowC，ColC)；
DecomposeCol(C，RowC，ColC)；
Row
C>>＝1；
Col
C>>＝1；
L＝L-1；
}
}
Wherein, function WaveletDecompose is the function that is used for input picture component C is decomposed picture breakdown level L.Function DecomposeRow and DecomposeCol are respectively the functions that is used at line direction and column direction decomposition input picture component.The logical process that function DecomposeRow realizes is for to decompose the input picture component line by line.Particularly; In the process that each row to the input picture component decomposes; At first each row with the input picture component is decomposed into strange position data and even position data; Through two lifting process (realizing) strange and even position data is upgraded then,, at last low frequency and high-frequency output are carried out change of scale (being realized by function S caleRow) to form the output of low frequency and high-frequency respectively by function F irstLiftRow and SecondLiftRow.
Particularly, the logical process of function DecomposeRow realization can be realized through following block:
DecomposeRow(C，Row
C，Col
C)
{
For(i＝0；i<Row
C；i++)
{
LazyRow(C，i，Col
C)；
FirstLiftRow(C，i，Col
C)；
SecondLiftRow(C，i，Col
C)；
ScaleRow(C，i，Col
C)；
}
}
Particularly, logical process that each row of input picture component is decomposed into strange position data and even position data is (as shown in Figure 3): with i line data C [i] [0], and C [i] [1], L C [Col
C-2], C [Col
C-1] even position data C [i] [0], C [i] [2], L C [Col
C-2] and strange position Elements C [i] [1], C [i] [3], L C [Col
C-1] presses following sequential storage at array C [i] [Col
C-1] in: C [i] [0], C [i] [2], LC [Col
C-2], C [i] [1], C [i] [3], LC [Col
C-1].For example, this process can be passed through to realize with minor function LazyRow:
LazyRow(C，i，Col
C)
{
Rearrange?the?data?C[i][0]～C[i][Col
C-1]as?the?following?sequence：
C[i][0]，C[i][2]，
L?C[i][Col
C-2]，C[i][1，C[i][3]，
LC[i][Col
C-1]
}
Particularly, through two lifting process strange and even position data is upgraded to form the output of low frequency and high-frequency respectively.Wherein, the logical process that promotes for the first time is: at first calculate strange position data (high frequency coefficients) according to even position data, utilize the strange position data of upgrading (high frequency coefficients) to calculate even position data (low frequency coefficients) then.The logical process that promotes for the second time is with for the first time similar, but it is different to promote parameter.For example, these two lifting process can be realized (wherein, the process of FirstLiftRow is as shown in Figure 4) through function F irstLiftRow and SecondLiftRow respectively:
FirstLiftRow(C，i，Col
C)
{
NC＝Col
C/2；NP＝Col
C；
For(n＝0；n<NC-1；n++)
C[i][NC+n]＝C[i][NC+n]+alpha*(C[i][n]+C[i][n+1])；
C[i][NP-1]＝C[i][NP-1]+alpha*(C[i][NC-1]+C[i][NC-1])；
C[i][0]＝C[i][0]+beta*(C[i][NC]+C[i][NC])；
For(n＝1；n<NC；n++)
C[i][n]＝C[i][n]+beta*(C[i][NC+n]+C[i][NC+n-1])；
}
SecondLiftRow(C，i，Col
C)
{
NC＝Col
C/2；NP＝Col
C；
For(n＝0；n<NC-1；n++)
C[i][NC+n]＝C[i][NC+n]+gamma*(C[i][n]+C[i][n+1])；
C[i][NP-1]＝C[i][NP-1]+gamma*(C[i][NC-1]+C[i][NC-1])；
C[i][0]＝C[i][0]+dlta*(C[i][NC]+C[i][NC])；
For(n＝1；n<NC；n++)
C[i][n]＝C[i][n]+dlta*(C[i][NC+n]+C[i][NC+n-1])；
}
Particularly, output is carried out the logical process of change of scale and is to low frequency and high-frequency: the low frequency coefficients for twice lifting operation multiply by r, and for the high frequency coefficients of twice lifting operation divided by r.For example, this process can realize through function S caleRow:
ScaleRow(C，i，Col
C)
{
NC＝Col
C/2；
For(n＝0；n<NC；n++)
{
C[i][n]＝C[i][n]*r；
C[i][NC+n]＝C[i][NC+n]/r；
}
}
Wherein, r=1.1496043988602, alpha=-1.586134342059924, beta=-0.052980118572961, gamma=0.882911075530934, dlta=0.443506852043971.
Function DecomposeCol is the function that is used for decomposing at column direction the input picture component.The logical process that function DecomposeRow realizes decomposes for the input picture component being pursued row.Particularly, except the input picture component being pursued the row processing rather than handling line by line, DecomposeCol is basic identical for function DecomposeRow and function.For example, can function DecomposeCol be defined as follows:
DecomposeCol(C，Row
C，Col
C)
{
For(j＝0；j<Col
C；j++)
{
LazyCol(C，Row
C，j)；
FirstLiftCol(C，Row
C，j)；
SecondLiftCol(C，Row
C，j)；
ScaleCol(C，Row
C，j)；
}
}
Wherein, four subfunctions among the DecomposeCol are identical with the respective function among the DecomposeRow, except the input picture component being pursued row are handled rather than handling line by line.
Step S106, according to input picture important low-limit frequency wavelet coefficient select normalization factor.With the institute of input picture is important all decompose picture breakdown level L after, at first calculate the maximum value of the low-limit frequency coefficient block of each component.Suppose, with input picture component C (Row
COK * Col
CRow) decompose picture breakdown level L, then exporting Wavelet Coefficient Blocks is W
C(Row
COK * Col
CRow).In this piece, Row
C/ M is capable * Col
C(upper left hand block of M=pow (2, L) be given down-sampling rate) is the low-limit frequency piece to/M row, uses
Expression.Through this low-limit frequency piece
Can calculate the maximum value of wavelet coefficient in this low-limit frequency piece
Maximum value according to each input picture component C
Can be by following calculating normalization factor NF:
Wherein
Be illustrated between each component C and get
Maximum.
Step S108 normalizes to time domain with the low-limit frequency wavelet coefficient of each component.Particularly, the logical process of this step is: each wavelet coefficient in the low-limit frequency coefficient block of each component at first takes absolute value, and then multiply by 255.0 backs divided by normalization factor NF, at last round again.For example, this process can be passed through to realize with minor function:
NormalizeCoff(
)
{
For(i＝0；i<Row
C/M；i++)
For(j＝0；j<Col
C/M；j++)
{
}
}
This function carries out normalization according to normalization factor NF to the low-limit frequency piece of each member, thereby has guaranteed that normalization coefficient is positioned in [0,255].Wherein, Floor (float num) function is used to obtain the maximum integer that is not more than num.
Fig. 5 shows the block diagram according to the image down sampling system of the embodiment of the invention that is used to realize said method.As shown in Figure 5, this image down sampling system comprises: decompose level and confirm unit 502, be used for confirming the picture breakdown level according to predetermined image down sampling rate; Wavelet decomposition performance element 504 is used for according to determined picture breakdown level each component of input picture being carried out wavelet decomposition; Normalization factor is confirmed unit 506, is used for the low-limit frequency wavelet coefficient according to each component after the wavelet decomposition, confirm to be used for to after the wavelet decomposition of input picture important low-limit frequency wavelet coefficient carry out normalized normalization factor; And component normalization performance element 508, be used for according to determined normalization factor after with the wavelet decomposition of input picture important low-limit frequency wavelet coefficient normalize to time domain.
It will be appreciated by those skilled in the art that; In certain embodiments; The function of said system and method can use hardware or firmware components (for example, application-specific integrated circuit (ASIC) (ASIC), erasable removing and programmable read-only memory (EEPROM) etc.) or other relevant assembly of pre-programmed to realize.In other embodiments, the function of said system and method can use a series of a plurality of calculation element to accomplish.Wherein, those calculation elements can access stored be useful on the code memory (not shown) of the computer readable program code of this calculation element of operation.Computer readable program code can be stored in fixing, tangible and the medium that can directly read by these assemblies on (for example; Dismountable disk, CD-ROM, ROM, hard disk, USB drive); Perhaps computer readable program code can still can be sent to these assemblies via the modulator-demodulator that is connected to network (including but not limited to the Internet) or other interface equipment through transmission medium by remotely storage.Transmission medium can be non-wireless medium (for example, light or analog communication line) or wireless medium (for example, microwave, infrared ray, free-space optical systems or other transmission plan) or their combination.
It will be understood by those skilled in the art that also to have more how optional execution mode and the improved procedure that can be used for realizing the embodiment of the invention, and above-mentioned execution mode and example only are the explanations of one or more embodiment.Therefore, scope of the present invention is only limited by appended claims.
Claims (10)
1. an image down sampling method is characterized in that, may further comprise the steps:
Confirm the picture breakdown level according to predetermined image down sampling rate;
According to said picture breakdown level each component of input picture is carried out wavelet decomposition;
Low-limit frequency wavelet coefficient according to each component after the wavelet decomposition; Confirm to be used for to after the wavelet decomposition of said input picture important low-limit frequency wavelet coefficient carry out normalized normalization factor; Wherein, with after the wavelet decomposition of said input picture the maximum value of important low-limit frequency wavelet coefficient as said normalization factor; And
According to said normalization factor after with the wavelet decomposition of said input picture important low-limit frequency wavelet coefficient normalize to time domain.
2. image down sampling method according to claim 1 is characterized in that, confirms said picture breakdown level: L=log through following formula
2(M), wherein, M representes said image down sampling rate, and L representes said picture breakdown level, log
2Expression is the logarithm at the end with 2.
3. image down sampling method according to claim 1 and 2 is characterized in that, adopts biorthogonal Daubechies 9/7 tap filter to come each component of said input picture is carried out wavelet decomposition.
4. image down sampling method according to claim 1 and 2; It is characterized in that, be used for image down sampling rate that each component to said input picture carries out the horizontal direction down-sampling and be used for the image down sampling rate that each component to said input picture carries out the vertical direction down-sampling and equate.
5. image down sampling method according to claim 4; It is characterized in that; Earlier each component of said input picture is carried out the wavelet decomposition of horizontal direction, again the horizontal direction wavelet decomposition result of each component of said input picture is carried out the wavelet decomposition of vertical direction.
6. an image down sampling system is characterized in that, comprising:
Decompose level and confirm the unit, be used for confirming the picture breakdown level according to predetermined image down sampling rate;
The wavelet decomposition performance element is used for according to said picture breakdown level each component of input picture being carried out wavelet decomposition;
Normalization factor is confirmed the unit; Be used for low-limit frequency wavelet coefficient according to each component after the wavelet decomposition; Confirm to be used for to after the wavelet decomposition of said input picture important low-limit frequency wavelet coefficient carry out normalized normalization factor; Wherein, after the wavelet decomposition of said input picture the maximum value of important low-limit frequency wavelet coefficient be confirmed as said normalization factor; And
Component normalization performance element, be used for according to said normalization factor after with the wavelet decomposition of said input picture important low-limit frequency wavelet coefficient normalize to time domain.
7. image down sampling according to claim 6 system is characterized in that, said decomposition level confirms that the unit confirms said picture breakdown level: L=log through following formula
2(M), wherein, M representes said image down sampling rate, and L representes said picture breakdown level, log
2Expression is the logarithm at the end with 2.
8. according to claim 6 or 7 described image down sampling systems, it is characterized in that said wavelet decomposition performance element adopts biorthogonal Daubechies 9/7 tap filter to come each component of said input picture is carried out wavelet decomposition.
9. according to claim 6 or 7 described image down sampling systems; It is characterized in that, be used for image down sampling rate that each component to said input picture carries out the horizontal direction down-sampling and be used for the image down sampling rate that each component to said input picture carries out the vertical direction down-sampling and equate.
10. image down sampling according to claim 9 system; It is characterized in that; Said wavelet decomposition performance element carries out the wavelet decomposition of horizontal direction earlier to each component of said input picture, the horizontal direction wavelet decomposition result of each component of said input picture is carried out the wavelet decomposition of vertical direction again.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200810168487 CN101686389B (en) | 2008-09-28 | 2008-09-28 | Image down sampling method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200810168487 CN101686389B (en) | 2008-09-28 | 2008-09-28 | Image down sampling method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101686389A CN101686389A (en) | 2010-03-31 |
| CN101686389B true CN101686389B (en) | 2012-10-03 |
Family
ID=42049303
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200810168487 Expired - Fee Related CN101686389B (en) | 2008-09-28 | 2008-09-28 | Image down sampling method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101686389B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105227964B (en) * | 2014-06-03 | 2018-11-06 | 深圳先进技术研究院 | Video-frequency identifying method and system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1381142A (en) * | 2000-05-18 | 2002-11-20 | 皇家菲利浦电子有限公司 | Encoding method for compression of video sequence |
| US6728406B1 (en) * | 1999-09-24 | 2004-04-27 | Fujitsu Limited | Image analyzing apparatus and method as well as program record medium |
| CN101076117A (en) * | 2006-05-16 | 2007-11-21 | 索尼株式会社 | Image processing apparatus and image processing method |
-
2008
- 2008-09-28 CN CN 200810168487 patent/CN101686389B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6728406B1 (en) * | 1999-09-24 | 2004-04-27 | Fujitsu Limited | Image analyzing apparatus and method as well as program record medium |
| CN1381142A (en) * | 2000-05-18 | 2002-11-20 | 皇家菲利浦电子有限公司 | Encoding method for compression of video sequence |
| CN101076117A (en) * | 2006-05-16 | 2007-11-21 | 索尼株式会社 | Image processing apparatus and image processing method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101686389A (en) | 2010-03-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US6137914A (en) | Method and format for storing and selectively retrieving image data | |
| JP3800551B2 (en) | Data processing apparatus and method | |
| Roos et al. | Reversible intraframe compression of medical images | |
| Sweldens | Lifting scheme: a new philosophy in biorthogonal wavelet constructions | |
| US6873734B1 (en) | Method and apparatus for compression using reversible wavelet transforms and an embedded codestream | |
| JP3461821B2 (en) | Memory management system and memory management method | |
| US7492955B2 (en) | Method and apparatus for compression using reversible wavelet transforms and an embedded codestream | |
| US7277596B2 (en) | Apparatus configured to eliminate image data show-through | |
| US8218908B2 (en) | Mixed content image compression with two edge data representations | |
| US20110052088A1 (en) | High dynamic range image mapping with empirical mode decomposition | |
| US8031951B2 (en) | 2-dimensional signal encoding/decoding method and device | |
| JP2006141018A (en) | Image encoding using compression adjustment based on dynamic buffer capacity level | |
| CN101686389B (en) | Image down sampling method and system | |
| EP1229738B1 (en) | Image decompression from transform coefficients | |
| US7630568B2 (en) | System and method for low-resolution signal rendering from a hierarchical transform representation | |
| CN107481193A (en) | A kind of image interpolation method based on wavelet transformation | |
| Maske et al. | Comparison Of Image Compression Using Wavelet For Curvelettransform & Transmission Over Wireless Channel | |
| JP2001223888A (en) | Picture processing method, picture processor and recording medium | |
| US20050123210A1 (en) | Print processing of compressed noisy images | |
| CN104683818A (en) | Image compression method based on biorthogonal invariant set multi-wavelets | |
| US20040179745A1 (en) | Method and arrangement for encoding and decoding images | |
| US8073267B2 (en) | Apparatus and method for transforming between discrete cosine transform coefficient and discrete wavelet transform coefficient | |
| US20040131260A1 (en) | Method and apparatus for support and/or conversion of two image formats | |
| US20040090441A1 (en) | Method and device for translating two-dimensional data of a discrete wavelet transform system | |
| CN113706396B (en) | Remote sensing image noise reduction processing method based on sliding window function |
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: 20121003 Termination date: 20180928 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |