CN104200433B - The scan method and device of a kind of two-dimensional image wavelet transform - Google Patents
The scan method and device of a kind of two-dimensional image wavelet transform Download PDFInfo
- Publication number
- CN104200433B CN104200433B CN201410421122.8A CN201410421122A CN104200433B CN 104200433 B CN104200433 B CN 104200433B CN 201410421122 A CN201410421122 A CN 201410421122A CN 104200433 B CN104200433 B CN 104200433B
- Authority
- CN
- China
- Prior art keywords
- row
- scanned
- pixel
- order
- component
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 43
- 230000009466 transformation Effects 0.000 claims abstract description 84
- 238000006243 chemical reaction Methods 0.000 claims abstract description 41
- 238000001914 filtration Methods 0.000 description 17
- 238000010586 diagram Methods 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 5
- 230000006835 compression Effects 0.000 description 4
- 238000007906 compression Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 238000000354 decomposition reaction Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000010408 sweeping Methods 0.000 description 2
- 238000000844 transformation Methods 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 230000004888 barrier function Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000004807 localization Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000003739 neck Anatomy 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
Landscapes
- Compression Or Coding Systems Of Tv Signals (AREA)
- Compression Of Band Width Or Redundancy In Fax (AREA)
- Complex Calculations (AREA)
Abstract
The embodiment of the invention discloses a kind of scan method of two-dimensional image wavelet transform and device.The scan method of described image two-dimensional discrete wavelet conversion includes：In wavelet transformation data at the same level, it is scanned according to the secondary component described in preferential, row order the secondary ordered pair of row sequence in the low frequency component of wavelet transformation data；It is scanned according to the pixel described in preferential, row order the secondary ordered pair of row sequence in the horizontal component of wavelet transformation data；It is scanned according to the pixel described in preferential, row order the secondary ordered pair of row sequence in the vertical component of wavelet transformation data；The pixel in the diagonal components of the wavelet transformation data is scanned according to since the upper left corner, diagonally toward the scanning sequence for returning to the lower right corner.The scan method of two-dimensional image wavelet transform disclosed in the embodiment of the present invention and device reflect the directional characteristic of the different components of two-dimensional discrete wavelet conversion.
Description
Technical field
The present embodiments relate to technical field of image processing, more particularly to a kind of two-dimensional image wavelet transform is swept
Retouch method and apparatus.
Background technology
In classical signal analysis theory, Fourier's theory is a kind of analysis means most widely used, that effect is best.
But it is a kind of analysis method of pure frequency domain, it is impossible to provide the frequency information in local time's section.In recent years, wavelet transformation is made
For a kind of transform domain signal processing method, very fast development has been obtained.When wavelet transformation is solved well-frequency localization
Problem, by Pan and Zoom computing, realizes the analysis of the refinement to the different scale of signal, solving Fourier transformation can not
The technical barrier of solution.Therefore, wavelet transformation is in many necks such as signal analysis, image procossing, seismic prospecting and nonlinear science
Domain has obtained extensive utilization.
Due to the advantageous characteristic that wavelet transformation has in itself, it is also widely used in terms of image procossing.Fig. 1
Show the decomposition step for the two-dimensional image wavelet transform that prior art is provided.Referring to Fig. 1, to input picture 100
In one-level wavelet transformation, the row of row filtering low pass filter 101 and row filtering high-pass filter 102 to input picture is utilized respectively
Data are filtered, then filtered data are entered with down-sampled 103, the 104 of raw column data.Then it is low that row filtering is utilized respectively again
Bandpass filter 105,107 and row filtering high-pass filter 106,108 pairs it is down-sampled after data be filtered, and to filtered
Data enter down-sampled 109,110,111, the 112 of line data.So, the one-level small echo change for treating changing image is just completed
Change.The later image of changing image progress one-level wavelet transformation is treated to be divided into low frequency component 121, horizontal component 122, vertically divide
Amount 123 and diagonal components 124.
Fig. 2 shows the design sketch of two-dimensional image wavelet transform in the prior art.Referring to Fig. 2, in first order two dimension
After wavelet transform, original image is divided into first order low frequency component 201, first order horizontal component 202, the first order and vertically divided
Amount 203 and first order diagonal components 204.After second level two-dimensional discrete wavelet conversion is carried out, first order low frequency originally
Component 201 is divided into second level low frequency component 205, second level horizontal component 206, second level vertical component 207 and second again
Level diagonal components 208.
In obtained each component of one-level wavelet transformation is carried out to input picture 100, horizontal component be expert at filtering when by
Row filtering high pass filter filters, and when arranging filtering by row filtering low pass filter filtering, therefore it can more embody described defeated
Enter the contour of object of image 100 in the horizontal direction.Vertical component be expert at filtering when by go filtering low pass filter filtering, and
By row filtering high pass filter filters during row filtering, therefore it can more embody the thing of the input picture 100 in vertical direction
Body profile.Diagonal components, which are expert at, to be filtered and has row filtering high-pass filter when arranging filtering respectively and arrange filtering high-pass filter filter
Ripple, therefore it can more embody the contour of object of the input picture in the diagonal directions.
Image can regard the array being made up of the pixel being arranged in order between different row and columns as.And pixel is group
Into the elementary cell of image, it characterizes bright-dark degree of the image on the point representated by it using the gray scale of itself.By two dimension
View data after wavelet transform exists still in the form of two-dimensional array.But, in compression of images and image transmitting
During, it is necessary to above-mentioned two-dimensional array to be transformed to one-dimensional data, this is accomplished by the graded of view data one of two dimension
Scanning.And the compression ratio of compression of images will be directly influenced to the design of the scanning sequence of view data, and image transmitting
Efficiency of transmission.
Existing image scanning algorithm includes：Morton image scannings algorithm, Raster image scannings algorithm and Zig-
Zag image scanning algorithms.Fig. 3 shows the scanning sequence of Morton image scanning algorithms.Referring to Fig. 3, according to Morton images
Scanning algorithm, according to the other cell size of final conversion stage, image to be transformed is divided into 2 × 2 elementary area, in each figure
As between the pixel inside unit using first from left to right, then zigzag scanning from top to bottom, and different elementary areas it
Between also using first from left to right, then from top to bottom zigzag scanning.Fig. 4 shows the scanning time of Raster image scanning algorithms
Sequence.Referring to Fig. 4, during using Raster image scanning algorithms, to the view data after two-dimensional discrete wavelet conversion according to different
Component is scanned respectively, and between the pixel inside each component, is performed first from left to right, then sweeping from top to bottom
Retouch.Fig. 5 shows the scanning sequence of Zig-zag image scanning algorithms.Referring to Fig. 5, during using Zig-zag image scanning algorithms,
Rank and component are not differentiated between, using the diagonal scan from the upper left corner to the lower right corner.
Above-mentioned three kinds scan the directional characteristic for not accounting for components different after one-level wavelet transformation.Raster images are swept
Although retouching algorithm to be scanned different components respectively, but unified scan mode is taken to different components.Institute
So that above-mentioned three kinds of image scanning algorithms can not reflect the directional characteristic of different components after one-level wavelet transformation.
The content of the invention
In view of this, the embodiment of the present invention proposes the scan method and device of a kind of two-dimensional image wavelet transform, with
Reflect the directional characteristic of the different components of two-dimensional discrete wavelet conversion.
In a first aspect, the embodiments of the invention provide a kind of scan method of two-dimensional image wavelet transform, the side
Method is scanned in wavelet transformation data not at the same level according to the preferential order of senior wavelet transformation data, methods described bag
Include：
In wavelet transformation data at the same level, according to wavelet transformation data described in preferential, row order the secondary ordered pair of row sequence
Low frequency component in secondary component be scanned；
Swept according to the pixel described in preferential, row order the secondary ordered pair of row sequence in the horizontal component of wavelet transformation data
Retouch；
Swept according to the pixel described in preferential, row order the secondary ordered pair of row sequence in the vertical component of wavelet transformation data
Retouch；
According to since the upper left corner, diagonally toward pair of the scanning sequence to the wavelet transformation data for returning to the lower right corner
Pixel in linea angulata component is scanned.
Second aspect, the embodiments of the invention provide a kind of scanning means of two-dimensional image wavelet transform, the dress
Put in wavelet transformation data not at the same level and to be scanned according to the preferential order of senior wavelet transformation data, described device bag
Include：
Low frequency component scan module, in the wavelet transformation data of same one-level wavelet transformation according to row sequence is preferential, row
Order secondary ordered pair described in wavelet transformation data low frequency component in secondary component be scanned；
Horizontal component scan module, for according to wavelet transformation data described in preferential, row order the secondary ordered pair of row sequence
Pixel in horizontal component is scanned；
Vertical component scan module, for according to wavelet transformation data described in preferential, row order the secondary ordered pair of row sequence
Pixel in vertical component is scanned；
Diagonal components scan module, for according to since the upper left corner, diagonally toward the scanning time for returning to the lower right corner
Pixel in the diagonal components of wavelet transformation data described in ordered pair is scanned.
The scan method and device of two-dimensional image wavelet transform provided in an embodiment of the present invention pass through in the small of peer
In wave conversion data, according to the secondary described in preferential, row order the secondary ordered pair of row sequence in the low frequency component of wavelet transformation data
Component is scanned, according to the pixel described in preferential, row order the secondary ordered pair of row sequence in the horizontal component of wavelet transformation data
It is scanned, is carried out according to the pixel described in preferential, row order the secondary ordered pair of row sequence in the vertical component of wavelet transformation data
Scanning, according to since the upper left corner, diagonally toward returning to the scanning sequence in the lower right corner to the diagonal of the wavelet transformation data
Pixel in line component is scanned, so as to employ different scanning sides to the different components of two-dimensional image wavelet transform
Formula, reflects the directional characteristic of the different components of two-dimensional discrete wavelet conversion.
Brief description of the drawings
By reading the detailed description made to non-limiting example made with reference to the following drawings, of the invention is other
Feature, objects and advantages will become more apparent upon：
Fig. 1 is the decomposition step signal of one-level wavelet transformation in the two-dimensional image wavelet transform that prior art is provided
Figure；
Fig. 2 is the effect diagram for the two-dimensional image wavelet transform that prior art is provided；
Fig. 3 is the scanning sequence schematic diagram for the Morton image scanning algorithms that prior art is provided；
Fig. 4 is the scanning sequence schematic diagram for the Raster image scanning algorithms that prior art is provided；
Fig. 5 is the scanning sequence schematic diagram for the Zig-zag image scanning algorithms that prior art is provided；
Fig. 6 is the flow chart of the scan method for the two-dimensional image wavelet transform that first embodiment of the invention is provided；
Fig. 7 is the flow chart of the scan method for the two-dimensional image wavelet transform that second embodiment of the invention is provided；
Fig. 8 is that the scanning sequence of the scan method for the two-dimensional image wavelet transform that second embodiment of the invention is provided shows
It is intended to；
Fig. 9 is the flow chart of the scan method for the two-dimensional image wavelet transform that third embodiment of the invention is provided；
Figure 10 is the scanning sequence of the scan method for the two-dimensional image wavelet transform that third embodiment of the invention is provided
Schematic diagram；
Figure 11 is the structure chart of the scanning means for the two-dimensional image wavelet transform that fourth embodiment of the invention is provided.
Embodiment
The present invention is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is only used for explaining the present invention, rather than limitation of the invention.It also should be noted that, for the ease of
Description, part related to the present invention rather than full content are illustrate only in accompanying drawing.
Fig. 6 shows the first embodiment of the present invention.
Fig. 6 is the flow chart of the scan method for the two-dimensional image wavelet transform that first embodiment of the invention is provided.Institute
The scanning means of two-dimensional image wavelet transform is stated in wavelet transformation data not at the same level according to senior wavelet transformation data
Preferential order is scanned.Referring to Fig. 6, the scan method of described image two-dimensional discrete wavelet conversion includes：
S610, in wavelet transformation data at the same level, according to wavelet transformation described in preferential, row order the secondary ordered pair of row sequence
Secondary component in the low frequency component of data is scanned.
Two-dimensional discrete wavelet conversion to input picture is carried out according to different ranks.One is being carried out to input picture
After level two-dimensional discrete wavelet conversion, conversion data are divided into low frequency component, horizontal component, vertical component and diagonal point
Amount.After the first order two-dimensional discrete wavelet conversion to input picture is completed, the low frequency component in conversion data is carried out
Second level two-dimensional discrete wavelet conversion.
Because the two-dimensional discrete wavelet conversion to input picture is carried out according to different ranks, the scanning to input picture
Carried out according to different ranks.From unlike wavelet transform procedure, it is preferential to the scanning process of image in conversion process most
High level conversion data are scanned.That is, preferential each component to the conversion data of highest level is scanned,
Each component of conversion data other to lower level is scanned again.If for example, carrying out 2-d discrete wavelet to input picture
The two-dimensional discrete wavelet conversion of two-stage is only carried out during conversion to the input picture, then be scanned to conversion data
When, it should it is preferential that each component that the second level converts data is scanned, then each component of first order conversion data is entered
Row scanning.
When the conversion data to same one-level are scanned, the low frequency component in this grade of conversion data is swept first
Retouch.Also, when being scanned to the low frequency component, according to result data described in preferential, row order the secondary ordered pair of row sequence
Low frequency component in secondary component be scanned.That is, to the secondary component elder generation in the low frequency component from left to right,
Carry out zigzag scanning from top to bottom again.Wherein, the secondary component refers to the low frequency point of next stage two-dimensional discrete wavelet conversion
Amount, horizontal component, vertical component and diagonal components.
S620, according to the pixel described in preferential, row order the secondary ordered pair of row sequence in the horizontal component of wavelet transformation data
It is scanned.
After completing to the scanning of same one-stage transfor-mation data low frequency components, horizontal component is scanned.To level
When component is scanned, using preferential, row order the И font scanning sequencies of row sequence.Namely to the picture in horizontal component
Element, is first scanned according to row sequence, is scanned according still further to row sequence.
It is preferred that, all pixels in the horizontal component can be scanned according to row chain order.It is described to pixel
The scanning carried out according to row chain order is exactly that all pixels are carried out first from top to bottom, then scanning from left to right.Can also be by
The elementary area of sets of adjacent pixels squarely in the horizontal component, enters the scanning of ranks chain order between elementary area.
That is, some pixels adjacent in the horizontal component are constituted into an elementary area, then between described image unit
Carry out first from top to bottom, then scanning from left to right.Meanwhile, inside each elementary area between different pixels, also carry out
The scanning of row chain order, that is, first from top to bottom, then scanning from left to right.
It is further preferred that described image unit can be the elementary area of adjacent 2 × 2 pixels composition.
S630, according to the pixel described in preferential, row order the secondary ordered pair of row sequence in the vertical component of wavelet transformation data
It is scanned.
After completing to the scanning of horizontal component in same one-stage transfor-mation data, vertical component is scanned.To vertical
When component is scanned, using preferential, row order the scanning sequency of row sequence.Namely to the pixel in vertical component, first press
It is scanned, is scanned according still further to row sequence according to row sequence.
It is preferred that, all pixels in the vertical component can be scanned according to row chain order.It is described to pixel
The scanning carried out according to row chain order is exactly that all pixels are carried out first from left to right, then scanning from top to bottom.Can also be by
The elementary area of sets of adjacent pixels squarely in the vertical component, enters the scanning of every trade chain order between elementary area.
That is, adjacent some pixels in the vertical component are constituted into an elementary area, then different elementary areas it
Between perform first from left to right, then scanning from top to bottom.Meanwhile, inside each elementary area between different pixels, also enter
The scanning of every trade chain order, that is, first from left to right, then scanning from top to bottom.
S640, according to since the upper left corner, diagonal is toward returning to the scanning sequence in the lower right corner to the wavelet transformation data
Diagonal components in pixel be scanned.
Zig-zag scanning sequences are can be described as since the upper left corner, diagonally toward the scanning sequence for returning to the lower right corner, it is complete
Into after to the scanning of vertical component in conversion at the same level, the diagonal components in being converted to this grade are scanned.To described right
When linea angulata component is scanned, the pixel in the diagonal components is scanned according to zig-zag scanning sequences.Namely
Say, the pixel in the diagonal components is performed since the upper left corner, diagonally toward the scanning for returning to the lower right corner.
In the present embodiment, because the horizontal component in one-level wavelet transformation data more can embody image in the horizontal direction
Contour of object, to the horizontal component using preferential, row order the scan mode of row sequence；Because vertical component can more embody
The contour of object of image in vertical direction, to the vertical component using preferential, row order the scan mode of row sequence；And by
The contour of object of image in the diagonal directions can be more embodied in diagonal components, to the diagonal components using diagonally
Scan mode.Different scan modes is used by the different components to two-dimensional image wavelet transform, the present embodiment is carried
The scan method of the two-dimensional image wavelet transform of confession preferably reflects the direction of the different components of two-dimensional discrete wavelet conversion
Characteristic.
Because the scan method of described image two-dimensional discrete wavelet conversion can really reflect two-dimensional discrete wavelet conversion
Mathematical characteristic, therefore described image two-dimensional discrete wavelet conversion scan method can preferably be applied to based on 2-d wavelet become
The fields such as the digital image compression changed.
The present embodiment is by wavelet transformation data at the same level, according to described in preferential, row order the secondary ordered pair of row sequence
Secondary component in the low frequency component of wavelet transformation data is scanned, according to described in preferential, row order the secondary ordered pair of row sequence
Pixel in the horizontal component of wavelet transformation data is scanned, according to small echo described in preferential, row order the secondary ordered pair of row sequence
The pixel converted in the vertical component of data is scanned, according to since the upper left corner, diagonally toward returning to sweeping for the lower right corner
Retouch the pixel described in time ordered pair in the diagonal components of wavelet transformation data to be scanned, so as to two-dimensional image discrete wavelet transformer
The different components changed employ different scan modes, reflect the directional characteristic of the different components of two-dimensional discrete wavelet conversion.
Fig. 7 and Fig. 8 show the second embodiment of the present invention.
Fig. 7 is the flow chart of the scan method for the two-dimensional image wavelet transform that second embodiment of the invention is provided.Institute
The scan method of two-dimensional image wavelet transform is stated based on first embodiment of the invention, it is further, excellent according to row sequence
First, row order secondary ordered pair described in wavelet transformation data horizontal component in pixel be scanned including：To the level
Pixel in component is scanned according to row chain order；According to wavelet transformation number described in preferential, row order the secondary ordered pair of row sequence
According to vertical component in pixel be scanned including：Pixel in the vertical component is scanned according to row chain order.
Referring to Fig. 7, the scan method of described image two-dimensional discrete wavelet conversion includes：
S710, in wavelet transformation data at the same level, according to wavelet transformation described in preferential, row order the secondary ordered pair of row sequence
Secondary component in the low frequency component of data is scanned.
S720, is scanned to the pixel in the horizontal component according to row chain order.
When pixel in the horizontal component is scanned, to the pixel in the horizontal component according to row chain order
It is scanned.That is, to the pixel in the horizontal component according to first from top to bottom, then order from left to right is swept
Retouch.
S730, is scanned to the pixel in the vertical component according to row chain order.
When pixel in the vertical component is scanned, to the pixel in the vertical component according to row chain order
It is scanned.That is, to the pixel in the vertical component according to first from left to right, then order from top to bottom is swept
Retouch.
S740, according to since the upper left corner, diagonally toward returning to the scanning sequence in the lower right corner to the wavelet transformation number
According to diagonal components in pixel be scanned.
Fig. 8 is that the scanning sequence of the scan method for the two-dimensional image wavelet transform that second embodiment of the invention is provided shows
It is intended to.Referring to Fig. 8, the scan method of the two-dimensional image wavelet transform provided according to the present embodiment is scanned to image
When, all pixels in the horizontal component are entered with the scanning of ranks chain order, that is, first from top to bottom, then from left to right
Scanning.All pixels in the vertical component are entered with the scanning of every trade chain order, that is, first from left to right, from top to bottom
Scanning.
During due to being scanned in the present embodiment to the horizontal component and the vertical component, all pixels are uniformly entered
The scanning of ranks chain order or row chain order, that is, in the scanning of horizontal component and vertical component, first to a line or one
Row pixel is integrally scanned, then scans other a row or column pixel, and when being scanned to full line or column of pixels
Do not take into account that other pixels adjacent thereto, therefore the scan method of two-dimensional image wavelet transform that the present embodiment is provided is more
Suitable for the less image of detail content.
The present embodiment to the pixel in the horizontal component according to row chain order by being scanned, and to described vertical
Pixel in component is scanned according to row chain order, and the direction for not only reflecting the different components of two-dimensional discrete wavelet conversion is special
Property, and cause image scan method to be more suitable for the less image of detail content.
Fig. 9 and Figure 10 show the third embodiment of the present invention.
Fig. 9 is the flow chart of the scan method for the two-dimensional image wavelet transform that third embodiment of the invention is provided.Institute
The scan method of two-dimensional image wavelet transform is stated based on first embodiment of the invention, it is further, excellent according to row sequence
First, row order secondary ordered pair described in wavelet transformation data horizontal component in pixel be scanned including：By horizontal component
It is scanned between interior pixel composition elementary area, described image unit according to row chain order；According to row sequence is preferential, row order
Secondary ordered pair described in wavelet transformation data vertical component in pixel be scanned including：By the pixel groups in vertical component
Into elementary area, it is scanned between described image unit according to row chain order.
Referring to Fig. 9, the scan method of described image two-dimensional discrete wavelet conversion includes：
S910, in wavelet transformation data at the same level, according to wavelet transformation described in preferential, row order the secondary ordered pair of row sequence
Secondary component in the low frequency component of data is scanned.
S920, the pixel in horizontal component is constituted between elementary area, described image unit according to the progress of row chain order
Scanning.
When being scanned to the horizontal component, the adjacent pixel in the horizontal component is constituted into image list first
Member, is then scanned between described image unit according to row chain order.That is, to different in the horizontal component
Elementary area is according to first from top to bottom, then order from left to right is scanned.
Further, when being scanned to horizontal component, to the pixel inside same elementary area, also according to row
Chain order is scanned.Namely inside described image unit, to different pixels also according to first from top to bottom, then from left to right
Order be scanned.
It is preferred that, described image unit is the elementary area of adjacent 2 × 2 pixels composition.
S930, the pixel in vertical component is constituted between elementary area, described image unit according to the progress of row chain order
Scanning.
When being scanned to the vertical component, the adjacent pixel in the vertical component is constituted into image list first
Member, is then scanned between described image unit according to row chain order.That is, to different in the vertical component
Elementary area is according to first from left to right, then order from top to bottom is scanned.
Further, when being scanned to vertical component, to the pixel inside same elementary area, also according to row
Chain order is scanned.Namely inside described image unit, to different pixels also according to first from left to right, then from top to bottom
Order be scanned.
It is preferred that, described image unit is the elementary area of adjacent 2 × 2 pixels composition.
S940, according to since the upper left corner, diagonally toward returning to the scanning sequence in the lower right corner to the wavelet transformation number
According to diagonal components in pixel be scanned.
Figure 10 is the scanning sequence of the scan method for the two-dimensional image wavelet transform that third embodiment of the invention is provided
Schematic diagram.It is right when the scan method of the two-dimensional image wavelet transform provided according to the present embodiment is scanned referring to Figure 10
Pixel in horizontal component first constitutes different elementary areas, is swept between different elementary areas according to row chain order
Retouch, and be scanned between pixel different inside the same elementary area also according to row chain order.And to vertical point
When pixel in amount is scanned, elementary area is first constituted to the pixel in vertical component, between different elementary areas according to
It is scanned, and is carried out between pixel different inside the same elementary area also according to row chain order according to row chain order
Scanning.
During due to being scanned in the present embodiment to the horizontal component and the vertical component, first by adjacent pixel groups
Into elementary area, it is scanned according still further to the spatial relationship between different images unit.For example, horizontal component is being performed
, can be in units of a few row pixels be during the preferential scanning of row sequence, can be right in the preferential scanning of the row sequence for this few row pixel
Adjacent pixel carries out scanning successively in a column direction with a pixel.Therefore the two-dimensional image that the present embodiment is provided is discrete small
The scan method of wave conversion is more suitable for the more image of detail content.
The present embodiment is by the way that when being scanned to the horizontal component and the vertical component, adjacent pixel is constituted
Elementary area, and image is scanned according to the spatial relationship between different images unit, not only reflect two-dimensional discrete small
The directional characteristic of wave conversion difference component, and cause scanning result to be more suitable for embodying the detail content of image.
Figure 11 shows the fourth embodiment of the present invention.
Figure 11 is the structure chart of the scanning means for the two-dimensional image wavelet transform that fourth embodiment of the invention is provided.Ginseng
See Figure 11, the scanning means of described image two-dimensional discrete wavelet conversion includes：Low frequency component scan module 1110, horizontal component are swept
Retouch module 1120, vertical component scan module 1130 and diagonal components scan module 1140.
The low frequency component scan module 1110 is used in the wavelet transformation data of same one-level wavelet transformation according to row sequence
Preferentially, row order secondary ordered pair described in wavelet transformation data low frequency component in secondary component be scanned.
The horizontal component scan module 1120 is used for according to wavelet transformation described in preferential, row order the secondary ordered pair of row sequence
Pixel in the horizontal component of data is scanned.
The vertical component scan module 1130 is used for according to wavelet transformation described in preferential, row order the secondary ordered pair of row sequence
Pixel in the vertical component of data is scanned.
The diagonal components scan module 1140 is used for according to since the upper left corner, diagonally toward returning to the lower right corner
Scanning sequence is scanned to the pixel in the diagonal components of the wavelet transformation data.
It is preferred that, the horizontal component scan module 1120 includes：First row chain order scanning element 1121.
The first row chain order scanning element 1121 is used to enter the pixel in the horizontal component according to row chain order
Row scanning.
The vertical component scan module 1130 includes：The first row chain order scanning element 1131.
The first row chain order scanning element 1131 is used to enter the pixel in the vertical component according to row chain order
Row scanning.
It is preferred that, the horizontal component scan module 1120 includes：Secondary series chain order scanning element 1122.
The secondary series chain order scanning element 1122 is used to the pixel in horizontal component constituting elementary area, the figure
As being scanned between unit according to row chain order.
The vertical component scan module 1130 includes：Second row chain order scanning element 1132.
The second row chain order scanning element 1132 is used to the pixel in vertical component constituting elementary area, the figure
As being scanned between unit according to row chain order.
It is preferred that, pixel composition elementary area of the secondary series chain order scanning element 1122 in by horizontal component,
And when being scanned between described image unit according to row chain order, inside described image unit, pressed between different pixels
It is scanned according to row chain order；
Pixel composition elementary area of the second row chain order scanning element 1132 in by vertical component, described image
When being scanned between unit according to row chain order, inside described image unit, according to row chain order between different pixels
It is scanned.
It is preferred that, the elementary area that described image unit constitutes for 2 × 2 adjacent pixels.
The embodiments of the present invention are for illustration only, and the quality of embodiment is not represented.
Will be appreciated by those skilled in the art that above-mentioned each module of the invention or each step can use general meter
Calculate device to realize, they can be concentrated on single computing device, or are distributed in the network that multiple computing devices are constituted
On, alternatively, they can be realized with the executable program code of computer installation, so as to be stored in storage
Performed in device by computing device, they are either fabricated to each integrated circuit modules respectively or will be many in them
Individual module or step are fabricated to single integrated circuit module to realize.So, the present invention be not restricted to any specific hardware and
The combination of software.
Each embodiment in this specification is described by the way of progressive, what each embodiment was stressed be with
Same or analogous part between the difference of other embodiment, each embodiment mutually referring to.
The preferred embodiments of the present invention are the foregoing is only, are not intended to limit the invention, for those skilled in the art
For, the present invention can have various changes and change.It is all any modifications made within spirit and principles of the present invention, equivalent
Replace, improve etc., it should be included in the scope of the protection.
Claims (10)
1. a kind of scan method of two-dimensional image wavelet transform, the scan method of described image two-dimensional discrete wavelet conversion exists
It is scanned in wavelet transformation data not at the same level according to the preferential order of senior wavelet transformation data, it is characterised in that including：
In wavelet transformation data at the same level, according to the low of wavelet transformation data described in preferential, row order the secondary ordered pair of row sequence
Secondary component in frequency component is scanned；
It is scanned according to the pixel described in preferential, row order the secondary ordered pair of row sequence in the horizontal component of wavelet transformation data；
It is scanned according to the pixel described in preferential, row order the secondary ordered pair of row sequence in the vertical component of wavelet transformation data；
According to since the upper left corner, diagonally toward diagonal of the scanning sequence to the wavelet transformation data for returning to the lower right corner
Pixel in component is scanned.
2. according to the method described in claim 1, it is characterised in that：
Bag is scanned according to the pixel described in preferential, row order the secondary ordered pair of row sequence in the horizontal component of wavelet transformation data
Include：
Pixel in the horizontal component is scanned according to row chain order；
Bag is scanned according to the pixel described in preferential, row order the secondary ordered pair of row sequence in the vertical component of wavelet transformation data
Include：
Pixel in the vertical component is scanned according to row chain order.
3. according to the method described in claim 1, it is characterised in that：
Bag is scanned according to the pixel described in preferential, row order the secondary ordered pair of row sequence in the horizontal component of wavelet transformation data
Include：
Pixel in horizontal component is constituted and is scanned between elementary area, described image unit according to row chain order；
Bag is scanned according to the pixel described in preferential, row order the secondary ordered pair of row sequence in the vertical component of wavelet transformation data
Include：
Pixel in vertical component is constituted and is scanned between elementary area, described image unit according to row chain order.
4. method according to claim 3, it is characterised in that：
Between pixel composition elementary area, and described image unit in by horizontal component it is scanned according to row chain order
When, inside described image unit, it is scanned between different pixels according to row chain order；
Between pixel composition elementary area in by vertical component, described image unit it is scanned according to row chain order
When, inside described image unit, it is scanned between different pixels according to row chain order.
5. the method according to claim 3 or 4, it is characterised in that described image unit constitutes for 2 × 2 adjacent pixels
Elementary area.
6. a kind of scanning means of two-dimensional image wavelet transform, the scanning means of described image two-dimensional discrete wavelet conversion exists
It is scanned in wavelet transformation data not at the same level according to the preferential order of senior wavelet transformation data, it is characterised in that including：
Low frequency component scan module, in the wavelet transformation data of same one-level wavelet transformation according to row sequence is preferential, row are arranged in order
Secondary ordered pair described in wavelet transformation data low frequency component in secondary component be scanned；
Horizontal component scan module, for the level according to wavelet transformation data described in preferential, row order the secondary ordered pair of row sequence
Pixel in component is scanned；
Vertical component scan module, for according to the vertical of wavelet transformation data described in preferential, row order the secondary ordered pair of row sequence
Pixel in component is scanned；
Diagonal components scan module, for according to since the upper left corner, diagonally toward the scanning sequence pair for returning to the lower right corner
Pixel in the diagonal components of the wavelet transformation data is scanned.
7. device according to claim 6, it is characterised in that：
The horizontal component scan module includes：
First row chain order scanning element, for being scanned to the pixel in the horizontal component according to row chain order；
The vertical component scan module includes：
The first row chain order scanning element, for being scanned to the pixel in the vertical component according to row chain order.
8. device according to claim 6, it is characterised in that：
The horizontal component scan module includes：
Secondary series chain order scanning element, for the pixel in horizontal component to be constituted between elementary area, described image unit
It is scanned according to row chain order；
The vertical component scan module includes：
Second row chain order scanning element, for the pixel in vertical component to be constituted between elementary area, described image unit
It is scanned according to row chain order.
9. device according to claim 8, it is characterised in that：
The secondary series chain order scanning element in by horizontal component pixel composition elementary area, and described image unit it
Between be scanned according to row chain order when, inside described image unit, swept between different pixels according to row chain order
Retouch；
Between pixel composition elementary area of the second row chain order scanning element in by vertical component, described image unit
When being scanned according to row chain order, inside described image unit, it is scanned between different pixels according to row chain order.
10. device according to claim 8 or claim 9, it is characterised in that described image unit constitutes for 2 × 2 adjacent pixels
Elementary area.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410421122.8A CN104200433B (en) | 2014-08-25 | 2014-08-25 | The scan method and device of a kind of two-dimensional image wavelet transform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410421122.8A CN104200433B (en) | 2014-08-25 | 2014-08-25 | The scan method and device of a kind of two-dimensional image wavelet transform |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104200433A CN104200433A (en) | 2014-12-10 |
| CN104200433B true CN104200433B (en) | 2017-07-28 |
Family
ID=52085719
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410421122.8A Active CN104200433B (en) | 2014-08-25 | 2014-08-25 | The scan method and device of a kind of two-dimensional image wavelet transform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104200433B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1845734A2 (en) * | 2006-04-12 | 2007-10-17 | Xerox Corporation | Decompression with reduced ringing artifacts |
| WO2008003985A2 (en) * | 2006-07-07 | 2008-01-10 | Cambridge Enterprise Limited | Method of identifying linear features within an image |
| CN101316364A (en) * | 2008-07-15 | 2008-12-03 | 南京信息工程大学 | Image compression system |
| CN101493945A (en) * | 2009-03-13 | 2009-07-29 | 北京航空航天大学 | Point interference image compression method and device based on wavelet transform |
-
2014
- 2014-08-25 CN CN201410421122.8A patent/CN104200433B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1845734A2 (en) * | 2006-04-12 | 2007-10-17 | Xerox Corporation | Decompression with reduced ringing artifacts |
| WO2008003985A2 (en) * | 2006-07-07 | 2008-01-10 | Cambridge Enterprise Limited | Method of identifying linear features within an image |
| CN101316364A (en) * | 2008-07-15 | 2008-12-03 | 南京信息工程大学 | Image compression system |
| CN101493945A (en) * | 2009-03-13 | 2009-07-29 | 北京航空航天大学 | Point interference image compression method and device based on wavelet transform |
Non-Patent Citations (3)
| Title |
|---|
| The contourlet transform: An efficient directional multiresolution image representation;Do M N等;《IEEE Transactions on Image Processing》;20051231;第14卷(第12期);第2091-2106页 * |
| 基于小波变换的图像压缩;王涛等;《交通与计算机》;20040830;第22卷(第119期);第30-33页 * |
| 基于纹理的自适应提升小波变换图像压缩;张军等;《计算机学报》;20100131;第33卷(第1期);第184-192页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104200433A (en) | 2014-12-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Luo et al. | Pansharpening via unsupervised convolutional neural networks | |
| Leung et al. | Least-squares luma–chroma demultiplexing algorithm for Bayer demosaicking | |
| Zheng et al. | Multisource image fusion method using support value transform | |
| US6252577B1 (en) | Efficient methodology for scaling and transferring images | |
| JP2019117644A (en) | Image upscaling | |
| US20150036917A1 (en) | Stereo image processing device and stereo image processing method | |
| US9355434B2 (en) | Method and device for reconstructing super-resolution images | |
| CN109300096A (en) | A kind of multi-focus image fusing method and device | |
| CN101753799B (en) | Median filtering method and median filtering circuit using the same | |
| Sepas-Moghaddam et al. | Light field denoising: exploiting the redundancy of an epipolar sequence representation | |
| CN102831581A (en) | Method for reconstructing super-resolution image | |
| Zhou et al. | Multiframe super resolution reconstruction method based on light field angular images | |
| Hua et al. | Context-aware joint dictionary learning for color image demosaicking | |
| CN104217412B (en) | Airborne super-resolution image reconstruction device and reconstruction method | |
| Chen et al. | Super resolution from a single image based on self similarity | |
| CN104200433B (en) | The scan method and device of a kind of two-dimensional image wavelet transform | |
| Yoo et al. | 3D image reconstruction from multi-focus microscope: axial super-resolution and multiple-frame processing | |
| CN105654017B (en) | two-dimensional code transmission method and system | |
| JP5514132B2 (en) | Image reduction device, image enlargement device, and program thereof | |
| CN105631810A (en) | Printing method and printing system based on pixel gridding dot multiplexing | |
| Yeh et al. | VDNet: video deinterlacing network based on coarse adaptive module and deformable recurrent residual network | |
| CN112598577B (en) | Image interpolation method, system and storage medium based on dislocation sampling | |
| CN105469399B (en) | A kind of method for reconstructing and device of the human face super-resolution towards mixed noise | |
| Mokari et al. | Fast single image SR via dictionary learning | |
| Chen et al. | A study of efficient light field subsampling and reconstruction strategies |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| C53 | Correction of patent of invention or patent application | ||
| CB02 | Change of applicant information |
Address after: 100089 Beijing, West Third Ring Road North, No. 25, green building, room 701, Haidian District Applicant after: NORTHKING INFORMATION TECHNOLOGY CO.,LTD. Address before: 100089 Beijing, West Third Ring Road North, No. 25, green building, room 701, Haidian District Applicant before: Beijing Northking Information Technology Co.,Ltd. |
|
| COR | Change of bibliographic data |
Free format text: CORRECT: APPLICANT; FROM: BEIJING NORTH KING INFORMATION TECHNOLOGY CO., LTD. TO: NORTHKING INFORMATION TECHNOLOGY CO., LTD. |
|
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |