CN1178124C - Image data compressing circuit and compressing and decompressing method - Google Patents
Image data compressing circuit and compressing and decompressing method Download PDFInfo
- Publication number
- CN1178124C CN1178124C CNB991097580A CN99109758A CN1178124C CN 1178124 C CN1178124 C CN 1178124C CN B991097580 A CNB991097580 A CN B991097580A CN 99109758 A CN99109758 A CN 99109758A CN 1178124 C CN1178124 C CN 1178124C
- Authority
- CN
- China
- Prior art keywords
- old
- color data
- pixel
- row coordinate
- row
- 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
Landscapes
- Compression Or Coding Systems Of Tv Signals (AREA)
- Compression Of Band Width Or Redundancy In Fax (AREA)
Abstract
The present invention relates to an image data compression circuit and a compression / decompression method. The compression circuit comprises a photoelectric sensing mould train which comprises a plurality of picture element sensors used for picking up corresponding picture element data. Each picture element datum of the image data is divided into first color data to third color data. A signal transforming and processing unit is used for receiving a control signal and the image data and outputting the compressed image data; the 2n-1 row of the compressed image data comprises the first color data and partial second color data, the 2n row of the compressed image data comprises the first color data and partial third color data. The compression method outputs the corresponding color data of the picture element according to the coordinate of the picture element.
Description
The invention relates to a kind of image data compressor circuit and compressing/decompressing method, particularly about a kind of in scanning process, can compress image data to improve the video compressor circuit and the method for sweep velocity.
Along with the development of computer technology, the pattern that the user contacted enters into multimedia (Multi-Media) epoch by dull type mode.And multimedia means that data to be processed comprise contents such as literal, video, sound.And, developed now and the multiple multimedia device that picks up video or voice signal for data can be represented with the multimedia form.
As the video input media, scanner (Scanner) is present more common device with digital camera (DigitalCamera).With regard to scanner, platform-type scanner (FlatbedScanner) is a kind of more common scanner, it converts numerical data in order to some files to be scanned to as penetration file (as transparency) or reflective file (as photo or paper) etc. via scanning sequence.Generally, the used optoelectronic induction module of platform-type scanner mainly comprises a photovoltaic sensing element, this photovoltaic sensing element can be: charge coupled cell (Charge Coupled Device, CCD) or contact video inductor (Contact ImageSensor, CIS).Platform-type scanner is in order to will be positioned over manuscript to be scanned on the glass platform and convert digital data to through scanning process and to input in the computer, for further utilizing.
In addition, still develop a kind of multi-function processor (multi-functional peripheral is called for short MFP) that comprises functions such as fax, photomechanical printing, scanning at present.Can will be combined on the machine by multi-function processor, to reduce the occupied space of office as facsimile recorder, photoprinter and scanner etc.
In scanning process, video to be scanned reflects the light that light source sent or transmission, and this reflected light or transmitted light are received by the optoelectronic induction module.Afterwards, be connected to main frame through control circuit again,, handle and carry out video via PC such as computer.
It is that the three primary colors of light-red (R), blue (B) and green (G) can be obtained the various coloured light that have according to different ratio combinations.The various conversely speaking, combinations that have coloured light to can be considered R, G and B.Below, various have the composition of interior R, G that is comprised of coloured light and B to abbreviate red (R), green (G) and blue (B) data as.
In scanning process, the optoelectronic induction module picks up R, G and the B data of each pixel in the image data row to be scanned earlier, and produces corresponding simulating signal.Again these simulating signals given Digital to Analog Converter (analog-digital converter), handle for next unit to produce numerical data; Then the optoelectronic induction module picks up image data to be scanned for R, G and B data in the next column, repeats said procedure.In this process, can roughly be divided into two types: single channel scanning (one-channel scanning) and three channels (three-channel scanning) according to the difference of its scanning, deal with data mode.Single channel scanning is meant the interior scanning module of scanister when scan image, once only reads, handles the R data (or G data or B data) of pixel in file one row to be scanned; Three channel scannings then are meant scanning module when scan image, once can read, handle R data, G data and the B data of pixel in file one row to be scanned simultaneously.Three channel scannings can increase the speed of scanister scan image.This dual mode widely people in the industry has now, is not described in detail in this.The present invention can be applicable to the single channel scanning or the type of three channel scannings.
Shown in 1A and 1B, but regardless of being single channel scanning or three channel scannings all application table 1A and 1B type.
Table 1A represents how prior art selects scanning starting position and the scanning area of R, G, B.In table 1A, each grid is all represented a kind of color data of the pixel on the zone to be scanned; Numeral in the pixel is then represented the position of this pixel on zone to be scanned.Among the present invention, (wherein x represents the row coordinate to the image data-pixel value of pixel for x, y) expression, and y then represents capable coordinate (x, y 〉=1) with symbol P.Such as, P (2n-1,1) promptly represents the pixel value of the 1st pixel in 2n-1 (n the is a natural number) row.
In the prior art, when scanning 2n-1 row, the scanning starting position of R, G, B is all at the 1st pixel in zone to be scanned, and its scanning area is complete row.Just its resolution is complete.
Table 1B represents to produce corresponding to each pixel the situation of corresponding R, B, G data.In table 1 B, R (x, y), G (x, y), (x y) represents pixel value P (x, y) included color data respectively to B.If with 8 bits (bit) expression, then the data volume that each pixel produced is 3 * 8=24 bit respectively for R, the G of each pixel, B data.
Can find out that by table 1A and 1B existing scanner all sends out the R in each pixel, G, B data when transmitting image data to computer.
Yet the speed bottle-neck of existing scanister (or MFP) often occurs in scanister (or MFP) data is carried toward computer, so in order to add fast scan speed, reduce the data volume that transmits toward computer and will be effective method.Reduce data volume, compression will be a unique method.
Consider the characteristic of scanister, desirable data compression/decompression compression method need meet following properties:
(1) fixing compressibility: be applied in the graphics data compression technology of scanner existing, when figure to be scanned was simple, compression effectiveness better, the amount of digital data that produces is less.And when figure to be scanned was complicated, compression effectiveness amount of digital data relatively poor, that produce was more.Usually in zone to be scanned, have that the subregion figure is complicated, the subregion picture is more simple, when the scanning complex figure, because its compression effectiveness is relatively poor, therefore the data volume that passes past main frame increases suddenly, causes the time of host receiving data elongated, and the speed that receives data is slack-off.So must make the sweep velocity of scanner slack-off this moment, could make the PC inbound pacing can cooperate sweep velocity.If utilize prior art, can reduce sweep velocity, must reduce the translational speed of optoelectronic induction module.Because the optoelectronic induction module has inertia when moving, therefore in the process that slows down, may have the part image data and pick up less than, row that the optoelectronic induction module must fall back and just not pick up as yet to pick up again this moment.This way tends to influence quality of scanning, and can reduce overall scan speed.So, be applicable to that the ideal data compression method of scanner preferably adopts fixedly compressibility.
(2) low calculation complexity: carrying out the data compression purpose is to add fast scan speed.And after with data compression, must decompress could be with reduction of data, to handle.So if the algorithm that decompresses is too complicated, then spending will be longer in the last time of decompression.Like this, the long decompression time will be offset and be reduced the time that data volume is saved.And make overall scan speed slack-off.
The object of the present invention is to provide a kind of in scanning process, the effective image data compression set of packed data.
Another object of the present invention is to provide a kind of image data compressing/decompressing method of effectively packed data, method of the present invention meets fixedly compressibility and low requirement of calculating complexity.
For the present invention that achieves the above object takes following measure:
A kind of image data compressor circuit of the present invention in order to export compression image data to a main frame, comprising:
One optoelectronic induction module comprises at least two pixel sensors, and in order to pick up corresponding pixel data on the image data, each pixel data of this image data resolves into one first color data, one second color data and one the 3rd color data; And
One signal conversion processes unit, in order to accepting a control signal and image data, and output compression image data;
Wherein, the 2n-1 row of compression image data comprise first color data and part second color data, and the 2n row of compression image data comprise first color data and part the 3rd color data.
Wherein, described optoelectronic induction module is a charge coupled cell.
Wherein, described optoelectronic induction module is a contact video inductor.
Wherein, described signal conversion processes unit comprises:
One analog/digital signal conversion device receives and changes the output signal of described optoelectronic induction module;
One digital signal processing unit receives the output signal of described analog/digital signal conversion device and the control signal of described control module.
Wherein, more comprise:
The one scan control module is to provide control signal;
One exports buffer cell, connects the signal output part of described signal conversion processes unit;
One interface circuit is connected between output buffer cell and described main frame.
The decompression method of a kind of image data of the present invention, can have capable pixel value P (x of M row * N to one, y) the image data that is compressed decompresses, definition x is row coordinate (x=1～M), y for the row coordinate (y=1～N), this known pixel and color data thereof that is compressed that image data comprises is: (a) row coordinate x, and row coordinate y is when being arbitrary value, (x y) comprises one first color data C1 to pixel value P
Old(x, y); (b) row coordinate x is an odd number, and row coordinate y is when being odd number, and (x y) more comprises one second color data C2 to pixel value P
Old(x, y); And (c) row coordinate x is an even number, and row coordinate y is when being even number, and (x y) more comprises one the 3rd color data C3 to pixel value P
Old(x, y), this video decompression method can (x y) produces the second color data C2 that is short of at first pixel value P
New(x, y) and/or the 3rd color data C3
New(x, y), decompression method may further comprise the steps:
If row coordinate x is odd number, and row coordinate y is when being even number, pixel value P (x, second color data C2 y)
New(x is y) by C1
Old(x, y), C2
Old(x, y-1), C1
Old(x, y-1), C2
Old(x, y-1), C1
Old(x, y+1) institute's decision jointly;
If row coordinate x is even number, and row coordinate y is when being odd number, pixel value P (x, second color data y), C2
New(x is y) by C1
Old(x, y), C2
Old(x-1, y), C1
Old(x-1, y), C2
Old(x+1, y), C1
Old(x+1, y) institute's decision jointly;
If row coordinate x is odd number, and row coordinate y is when being even number, pixel value P (x, the 3rd color data C3 y)
New(x is y) by C1
Old(x, y), C3
Old(x-1, y), C1
Old(x-1, y), C3
Old(x+1, y), C1
Old(x+1, y) institute's decision jointly;
If row coordinate x is even number, and row coordinate y is when being even number, pixel value P (x, second color data C2 y)
New(x is y) by C1
Old(x, y), C1
Old(x-1, y-1), C1
Old(x+1, y-1), C1
Old(x-1, y+1), C1
Old(x+1, y+1), C2
Old(x-1, y-1), C2
Old(x+1, y-1), C1
Old(x-1, y+1), C2
Old(x+1, y+1) institute's decision jointly;
If row coordinate x is odd number, and row coordinate y is when being odd number, pixel value P (x, the 3rd color data C3 y)
New(x is y) by C1
Old(x, y), C1
Old(x-1, y-1), C1
Old(x+1, y-1), C1
Old(x-1, y+1), C1
Old(x+1, y+1), C3
Old(x-1, y-1), C3
Old(x+1, y-1), C3
Old(x-1, y+1), C3
Old(x+1, y+1) institute's decision jointly.
The compression method of a kind of image data of the present invention, can be compressed an image data with capable pixel of M row * N, and corresponding to each the pixel P among the image data (x, y), the color constituent of each pixel all comprises one first color data C1 (x, y), one second color data C2 (x, y) and one the 3rd color data C3 (x, y), at each pixel P (x, y), the compression method of image data is:
If the row coordinate x of this pixel is an odd number, when row coordinate y is odd number, export this pixel the first color data C1 (x, y) and the second color data C2 of this pixel (x, y);
If this pixel column coordinate x is odd number, when row coordinate y is even number, export this pixel the first color data C1 (x, y);
If this pixel column coordinate x is even number, when row coordinate y is odd number, export this pixel the first color data C1 (x, y);
If this pixel column coordinate x is even number, when row coordinate y is even number, export this pixel the first color data C1 (x, y) and the 3rd color data C3 of this pixel (x, y).
Conjunction with figs., table and embodiment are described in detail as follows circuit of the present invention and method:
The simple declaration of accompanying drawing, table:
Fig. 1: the circuit block diagram of first embodiment of scanister of the present invention;
Fig. 2: the circuit block diagram of second embodiment of scanister of the present invention;
Table 1A: the expression prior art is selected R, the G of each pixel in the zone to be scanned, the data of B;
Show 1B: after doing selection according to table 1A, the color data of image data pixel value;
Table 2A: a kind of R, G, B data of selecting each pixel in the zone to be scanned of expression first embodiment of the invention;
Table 2B: after doing selection according to table 2A, the color data of image data pixel value;
Table 3A: another of expression first embodiment of the invention selected each pixel R, G, B data in the zone to be scanned;
Table 3B: after doing selection according to table 3A, the color data of image data pixel value;
The table 4A: represent of the present invention another select R, the G of each pixel, the data of B;
Table 4B: after doing selection according to table 4A, the color data of image data pixel value.
As show shown in 2A and the 2B, select R, the G of each pixel in the zone to be scanned, the method synoptic diagram of B data in its expression first embodiment of the invention, the present invention treats the shades of colour data of each pixel in the scanning area and does suitable selection, to reach the purpose of compression image data, 2A can find out by table, and present embodiment is when scanning G data, send out normal mode scanning, that is scanning starting position is located at the 1st pixel in zone to be scanned, and its scanning area comprises all pixels, and its resolution is complete.Because for the optical sensor module (CCD or CIS) that extensively adopts at present, G is to the influence degree maximum of video brightness, so present embodiment keeps the G composition of each pixel, only R, B composition is compressed.
When the R data of scanning 2n-1 row, scanning starting position is located at the 2nd pixel of these row, then, scan this row with half scanning density (be exactly scanning area be half of permutation).That is the scanning area of R comprises that this is listed as all even number item pixels (the 2nd, 4 ... pixel etc.).So compared with the complete resolution of G color data, the resolution of R has only half.
Similarly, when the B data of scanning 2n row, scanning starting position is located at the 1st pixel of these row, scans this row with half scanning area then.That is the scanning area of B comprises that this is listed as all odd term pixels (the 1st, 3,5 ... pixel etc.).Equally, the resolution of B is also half as large.Symbol " X " among the table 2A represents this pixel not in the scanning area scope of R or B.
By table 2A as can be known, every row pixel data being done when selecting, is to select G, and R or B one of them.That is, shown in table 2A, when 2n-1 row (n is a natural number) are selected, be the G data of selecting all pixel values, and the R data of even number item pixel value.And the 2n row are G data of selecting all pixel values, and the B data of odd term pixel value.2n+1,2n+3 ... row are with identical selection mode with the 2n-1 row.And 2n+2,2n+4 ... row then adopt identical selection mode with the 2n row.
Certainly, the mode of above-mentioned selection B and R is interchangeable.Just, 2n-1,2n+1,2n+3 ... the scanning area of column selection G is complete, and the scanning area of B and resolution are half.2n, 2n+2,2n+4 ... the scanning area of column selection G is complete, and the scanning area of R and resolution are half.In addition, the beginning scanning position of R and B also can be all the 1st pixel and the 2nd pixel, and its scanning area comprises all even number item pixels or all odd term pixels.
As show shown in the 2B, its expression is according to the selection scanning starting position of table 2A and the method for resolution, and corresponding each pixel compresses the synoptic diagram of back data.In table 2B, each grid is represented a pixel.And data after compression are hereinafter referred to as packed data.
After using the inventive method, on average each pixel only transmits 3 * 8/2=12 bit and gives computer.And prior art need transmit 24 bits/pixel to computer.So application the present invention, compressibility can be 12/24=50%, and is a fixed value.This compressibility can compare just and can find out by showing 1B his-and-hers watches 2B.Because compressibility is fixed on 50%, so even use scanner of the present invention when the scanning complex figure, its compression speed also can be maintained fixed.So data compression method of the present invention as can be known can meet the compressibility of the 1st characteristic of good compression law regulation-fixedly.
When scanning process during with data compression, the computer part then need have corresponding decompression method that data are decompressed.Owing to when decompressing, must meet low calculation complexity, could improve overall scan speed.So the present invention proposes a kind of decompression method that meets low calculation complexity.The present invention is in the computer part, utilizes interpolation method (interpolation) with in the data compression process, and the partial data that is dropped is reduced when decompressing.
Below, when showing 2B to compress data into, a kind of interpolation method proposed by the invention is sketched.In following formula, in data compression process, the color data that is dropped is used R
New(2n-1, k+1), B
New(2n+1 k+2) waits expression, also can be referred to as unknown color data.As for originally not being compressed part in the pixel, then with as R
Old(2n-1, k), R
Old(2n k+1) represents it, also can be referred to as the known color data.In formula (1)～(4), n and k are all natural number.
The R of 2n-1 row
NewData ask method to be:
R
new(2n-1，k+1)＝G
old(2n-1，k+1)+[R
old(2n-1，k)-G
old(2n-1，k)+R
old(2n-1，k+2)-G
old(2n-1，k+2)]/2 (1)
Formula (1) the representative meaning is R
New(2n-1 is by G k+1)
Old(2n-1, k+1), R
Old(2n-1, k), G
Old(2n-1, k), R
Old(2n-1, k+2) and G
Old(2n-1, k+2) common decision.That is, obtain the mean value of (R-G) by left and right sides adjacent image point, obtain the R value by G+ (R-G) again.
In like manner, the B of 2n row
NewData ask method to be:
B
new(2n，k)＝G
old(2n，k)+[B
old(2n，k-1)-G
old(2n，k-1)+B
old(2n，k+1)-G
old(2n，k+1)]/2 (2)
The B of 2n-1 row
NewData ask method to be:
B
new(2n-1，k)＝G
old(2n-1，k)+[B
old(2n-2，k)-G
old(2n-2，k)+B
old(2n，k)-G
old(2n，k)]/2 (3)
The implication of formula (3) is, asks (B-G) mean value from pixel up and down, obtains the B value by G+ (B-G) again.
The R of 2n
NewData ask method to be:
R
new(2n，k)＝G
old(2n，k)+[R
old(2n-1，k)-G
old(2n-1，k)+R
old(2n+1，k)-G
old(2n+1，k)]/2 (4)
Aggregative formula (1)～(4) can be summarized, when if a certain pixel has unknown color data (R or B) to be decompressed, and when in the adjacent image point of this pixel (up and down or about) known color data of the same type being arranged, then should the unknown color data by the G data of this pixel with and the known color data of the same type of adjacent image point determine jointly.
Yet, when utilizing interpolation method to decompress, need consider boundary condition.Such as, when utilizing formula (3) to obtain the B of the 1st row
BewDuring data, B
New(1, k)=G
Old(1, k)+[B
Old(0, k)-G
Old(0, k)+B
Old(2, k)-G
Old(2, k)]/2, because shortcoming B
Old(0, k) and G
Old(0, k) can't obtain.Present embodiment is with B
Old(2, k) and G
Old(2, k) be considered as B
Old(0, k) and G
Old(0, k).
Below will discuss, and wait to decompress when a certain pixel has unknown color data, but the adjacent image point of this pixel there is no known color data of the same type, how obtain this unknown color data.
Formula (3) representative is to ask from pixel up and down, but can finds out by showing 2B when asking the B data of 2n-1 row, desires to ask B
New(2n-1 during k) value, need utilize its B data of two pixels: B up and down
New(2n-2, k) and B
New(2n, k).But B
New(2n-2, k) and B
New(2n k) still needs and utilizes formula (2) to obtain.So asking B
New(2n-1 in the time of k), is to utilize formula (2) to obtain B earlier
New(2n-2, k) and B
New(2n, value k) utilize formula (3) to try to achieve B again
New(2n-1, k).Can learn via deriving.
B
new(2n-1，k)＝G
old(2n-1，k)+[B
old(2n-2，k-1)-G
old(2n-2，k-1)+B
old(2n-2，k+1)-G
old(2n-2，k+1)+B
old(2n，k-1)-G
old(2n，k-1)+B
old(2n，k+1)-G
old(2n，k+1)]/4 (5)
The meaning of formula (5) representative is B
New(2n-1, value k) are by G
Old(2n-1, k), B
Old(2n-2, k-1), G
OlD (2n-2, k-1), B
Old(2n-2, k+1), G
Old(2n-2, k+1), B
Old(2n, k-1), G
Old(2n, k-1), B
Old(2n, k+1), G
Old(2n, k+1) common decision.
In like manner can demonstrate,prove, ask R
New(2n during k+1) value, utilizes formula (1) to obtain R earlier
New(2n-1, k+1) and R
New(2n+1, value k+1) utilize formula (4) to try to achieve again.
R
new(2n，k+1)＝G
old(2n，k+1)+[R
old(2n-1，k)-G
old(2n-1，k)+R
old(2n-1，k+2)-G
old(2n-1，k+2)+R
old(2n+1，k)-G
old(2n+1，k)+R
old(2n+2，k+2)-G
old(2n+2，k+2)]/4 (6)
The meaning of formula (6) representative is R
New(2n, value k+1) are by G
Old(2n, k+1), R
Old(2n-1, k), G
Old(2n-1, k), R
Old(2n-1, k+2), G
Old(2n-1, k+2), R
Old(2n+1, k), G
Old(2n+1, k), R
Old(2n+2, k+2) and G
Old(2n+2, k+2) institute's decision jointly.
Can summarize from formula (5) and (6), if B (or R) color data of a certain pixel value to be decompressed, and its adjacent image point does not up and down all have and comprises known color data of the same type, then need utilize the G chromatic number certificate of this pixel itself, with and the G chromatic number of upper left, upper right, lower-left and bottom right reach B (or R) chromatic number according to this and try to achieve jointly according to institute.
Because (R-G) of each pixel and value (B-G) are all very near 0 in the black and white video, so the R, B value that utilize each pixel that above-mentioned algorithm obtains are very near R, the B value of original value.So when utilizing above-mentioned interpolation method, resulting black and white video scanning effect also can be very good.Formula (1) to (4) is applicable to the packed data as table 2B.As for as described above, R and B exchange; Or the scanning starting position of R and B is all the situation of the 1st or the 2nd pixel, then can utilize the design of interpolation method to obtain corresponding formula.Then no longer repeat in this.
And observation can find behind formula (1)～(4) that this kind interpolation method only needs several simple additions, subtraction and 1 division arithmetic.So its calculation complexity is quite low.Can meet needed second characteristic, the one low calculation complexity of good compression method.Certainly, the person skilled in the art can derive from the design of above-mentioned formula and variously different interpolation methods and decompress.Above-mentioned only is an embodiment wherein.For example, also can use bicubic surface interpolation method (Bi-Cubic Interpolation), the value of (the B data-G data) of 16 pixels around utilizing, unknown data (B data or G data) that (R data-G data) try to achieve a certain pixel.
That is, select the scanning area of R and B to have four kinds of patterns.The scanning area that first kind of pattern is the R data of a certain row comprises all even number item pixels, and the scanning area of the B data of its adjacent column comprises all odd term pixels; The G data are complete scan then, shown in table 2A.Second kind of pattern then is: the R scanning area of a certain row comprises all even number item pixels, and the B scanning area of its adjacent column comprises all even number item pixels and the B scanning area of its adjacent column comprises all even number item pixels; The G data are complete scan then.The third pattern then is: the scanning area of a certain row R comprises all odd term pixels and the scanning area of its adjacent column B comprises all even number item pixels; The G data are complete scan then.The 4th kind of pattern then is: the scanning area of a certain row R comprises all odd term pixels and the B scanning area of its adjacent column comprises all even number item pixels; The G data are complete scan then.
Second asks the essence spirit of method to be similar to first kind of pattern to the unknown color data of four-mode.Just, when desiring to obtain a certain pixel unknown color data (R or B) being arranged, when in (up and down or about) known color data of the same type being arranged if the adjacent image point of this pixel, then should the unknown color data by the G data of this pixel with and the common decision of known color data institute of the same type of adjacent image point.When there is no known color data of the same type if the adjacent image point of this pixel in (up and down or about), then should the unknown color data by the common decision of known color data institute of the same type in upper left, the lower-left of the G data of this pixel and this pixel, upper right, the bottom right.
In the present embodiment, when selecting the scanning area of R, G, B, select as also can be shown in table 3A.As show shown in the 3A, when scanning 2n-1 row, the scanning area of G still is chosen as complete area.The selection of R and B then is all half zone.Be similar to the selection scanning area of table 2A, the scanning area of R also can comprise all even number item pixels (the 2nd, 4 ... pixel etc.) or all odd term pixels (the 1st, 3,5 ... pixel etc.).So the resolution of R has only half.And the scanning area of B comprises all odd term pixels or all even number item pixels too.So the resolution of B also has only half.It is each other independently that the scanning area of R is selected to select with the scanning area of B.
As show shown in the 3B, it is for comprising the synoptic diagram of data according to each pixel of table 3A gained.Each pixel all comprises 2 color data (one of them data of G data and R/B).So its compressibility is 2/3=67%.
Equally, be similar to table during 3B, also have four kinds of patterns available when image data is compressed into.First kind of pattern is respectively, and the scanning area of the R data of row comprises all even number item pixels; The B data scanning zone of each row comprises all odd term pixels; Every G data are complete scan then.As show shown in the 3A.Second kind of pattern then is: each scanning area that is listed as R comprises all even number item pixels; Each scanning area that is listed as B comprises all even number item pixels; The G data of each row are complete scan then.The third pattern then is: each scanning area that is listed as R comprises all odd term pixels; Each scanning area that is listed as B comprises all idol pixels; The G data of each row are complete scan then.The 4th kind of pattern then is: the R scanning area of each row comprises all odd term pixels; Each scanning area that is listed as B comprises all even number item pixels; The G data of each row are complete scan then.
Interpolation rule as for table 3B can be obtained compressed color data in the compression process with (2) or (3) and (4) by similar formula (1).So its detailed situation is no longer heavily chatted in this.
Image data is compressed into table 2B or the result of table behind the 3B, its advantage is respectively arranged.When being compressed into table 2B, its compression effectiveness is better.And be compressed into when showing the 3B state, its scanner distortion rate is lower.
In addition, though on tell bright to be to be that compression/de-compression is carried out on the basis with the adjacent image point, in fact also can to consider with a plurality of continuous pixels to be that unit carries out compression/de-compression.For example: if known R
Old(1, y) and R
Old(2, y), then it can be considered as last unit pixel.And if known R
Old(4, y) and R
Old(5, y), then it can be considered as back one unit pixel.Therefore, R
New(3, y)=G
Old(3, y)+1/4[R
Old(1, y)-G
Old(1, y)+R
Old(2, y-G
Old(2, y)+R
Old(4, y)-G
Old(4, y)+R
Old(5, y)-G
Old(5, y)].That is to say, merge the color data of the color data of a plurality of known pixels, at the unknown pixel that decompresses and be adjacent according to " average second color data of the Dan Su of unit with first color data difference " mutually as a unit pixel.
As shown in Figure 1, Fig. 1 represents the circuit block diagram of the first embodiment scanister 400.Wherein, scanister 400 comprises: and optoelectronic induction module 410, analog/digital signal conversion device (Analog/Digital Converter, ADC) 420, digital signal processing unit (DigitalSignal Processing Unit) 430, scan control unit (Scanning Control Unit) 440, output buffer (Output Buffer) 450 and interface (Interface) 460.
Optoelectronic induction module 410 comprises an optical inductor at least, this optical inductor can be charge coupled cell (Charge Coupled Device, CCD) or contact video inductor (ContactImage Sensor, CIS).When a certain image data of scanning, three kinds of data (R data, G data and B data) that this optical inductor will sense video convert simulating signal to, and are sent to next cell processing respectively.
When data are delivered to analog/digital signal conversion device 420, then be converted into digital data.The output signal of analog/digital signal conversion device 420 inputs to digital signal processing unit 430.
The control signal that digital signal processing unit 430 receives by 440 outputs of scan control unit.This control signal is in order to the resolution and the scanning area of gated sweep R, G, B data.Such as, when scanning R, resolution (half) and scanning starting position (the 2nd pixel) are controlled by control signal.That is image data (R, G, B data) compressed action is to carry out (that is choice of the R of each pixel, B color data) in digital signal processing unit 430.Selected resolution that goes out and scanning area are as mentioned above.
In circuit, analog/digital signal conversion device 420 and digital signal processing unit 430 can be combined, be called the analog/digital signal processing unit.
The output data of digital signal processing unit 430 is to input to output buffer 450.The output of output buffer 450 inputs to main frame by interface circuit 460, as PC 470.As mentioned above, the interpolation method calculating section that packed data is decompressed is to carry out in computer 470.
Image data can be compressed into shown in table 2A and 2B or table 3A and table 3B by scanister 400, and in computer 470, decompress.
Embodiments of the invention two there is no different with embodiment one in selecting to scan R data, G data and B data required scanning area, scanning starting position and resolution aspect.So present embodiment is selected the method for scanning starting position and resolution, shown in table 2A.Each pixel remaining data synoptic diagram as the table 2B shown in.With the also available formula of interpolation method (1)～(6) that packed data decompresses, no longer repeat.
As shown in Figure 2, it is present embodiment scanister 500 circuit block diagrams, wherein, scanister 500 comprises optoelectronic induction module 510, analog/digital signal conversion device 520, digital signal processing unit 530, scan control unit 540, compressing control unit 580, output buffer 550 and interface circuit 560.
Optoelectronic induction module 510 comprises optical inductor, and this optical inductor can be charge coupled cell or contact video inductor.When a certain image data of scanning, optoelectronic induction module 510 is together R data, G data and B data to be sent to next unit.
When data are delivered to analog/digital signal conversion device 520, then be converted into digital data.The output signal of analog/digital signal conversion device 520 inputs to digital signal processing unit 530.
Compressing control unit 580 receives the control signal of being exported by scan control unit 540.Though this control signal is in order to control resolution and scanning area, yet in this embodiment, the resolution of R, G, B data and scanning area all are complete.In other words, in the present embodiment, the image data compressed action does not carry out in digital signal processing unit 530.
In circuit, analog/digital signal conversion device 520 and digital signal processing unit 530 can be combined equally, be referred to as the analog/digital signal processing unit.
The output data of digital signal processing unit 530 inputs to compressing control unit 580.After a certain row image data (R, G, B data) entered into compressing control unit 580 inside, compressing control unit 580 decisions will be compressed which image data in the pixel.Its compression scenario is described as table 2A and 2B among first embodiment.Such as when scanning G, its scanning area is whole pixels.During scanning R/B, pixel is selected in its scanning area compartment of terrain, such as all odd term pixels or all even number item pixels.In other words, in the present embodiment, the video compression is to carry out in compressing control unit 580.
The output data of compressing control unit 580 inputs to main frame 570 by output buffer 550 and interface circuit 560, main frame (as PC).As above-mentioned, the calculating of the interpolation method that packed data is decompressed is to carry out in computer 570.This interpolation method can use formula (1) to (6).Certainly, decompress with interpolation method, above-mentioned only is an example wherein.
Scanister 500 can be compressed into image data as table 2A with shown in table 2B or table 3A and the table 3B.
Second embodiment and the first embodiment difference are: because analog/digital signal conversion device 420 after R data, G data and the B data of first in regular turn that this photovoltaic sensing element is the received pixel are sent respectively earlier, is sent R data, G data and the B data of second pixel again.Therefore, the optoelectronic induction module of present embodiment is all exported the image data (R, G, B data) of each pixel, and then compresses and (just to R, the G of each pixel, the choice of B data, as cast out B (2n-1, k+1) data).In other words, compression is changed in the compressing control unit 580 and is carried out.
Certainly, the mode that image color is decomposed does not only resolve into this a kind of method of light three primary colors, can be broken down into the printing three primary colors yet: blue-green (Cyan, C), aubergine (Magenta, M) and yellow (Yellow, Y).In addition, (z) etc. mode reaches identical purpose for x, y also color can be resolved into color three coordinates.
Compared with prior art, the present invention has following effect:
The present invention can be applicable to have on the scanister (as scanner or facsimile recorder) or multi-function processor of scan function.
Therefore, one of effect of the present invention is, in the scanning process, reaches the purpose of image data compression and raising overall scan speed with the scanning area (resolution) of gated sweep R, G, B data.Data after the compression are to utilize interpolation method that it is decompressed on computers.Like this, can obtain the fixedly compressibility of 50% (or 67%).Owing to fixedly compressibility is arranged, even when the scanning complex pattern, also can not make the sweep velocity step-down.And when the scanning black-white video, also can get and well be scanned effect.In addition, general existing scanister is many can to scan (not necessarily to whole file scan) at a certain regional area of file according to user's demand, and the present invention also can be applicable to this kind occasion certainly, states clearly hereby.
Another effect of the present invention is, utilizes simple and easy decompression method to improve overall scan speed.
In addition, in design scope of the present invention, also has different variation operational modes.For example, when selecting the scanning area of R, G, B, also can as showing shown in the 4A, select.As show shown in the 4A, when scanning 2n-1 row, the scanning area of G and R still is chosen as complete area.When scanning 2n row, the scanning area of G and B still is chosen as complete area.Certainly, R and B are interchangeable.
As show shown in the 4B, its expression comprises the synoptic diagram of data according to each pixel of table 4A gained.Wherein, each pixel all comprises 2 color data (data among G data and the R/B).So its compressibility is 2/3=67%.Its decompression mode is identical with above-mentioned other embodiment, is not described in detail in this.
More than narration is to borrow preferred embodiment that architectural feature of the present invention is described, is not to be used to limit protection scope of the present invention.
Table 1A
Table 1B
Table 2A
Table 2B
| G(2n-1，k) R(2n-1，k) | G(2n-1，k+1) | G(2n-1，k+2) R(2n-1，k+2) |
| G(2n，k) | G(2n，k+1) B(2n，k+1) | G(2n，k+2) |
| G(2n+1，k) R(2n+1，k) | G(2n+1，k+1) | G(2n+1，k+2) R(2n+1，k+2) |
Table 3A
Table 3B
| G(2n-1，k) B(2n-1，k) | G(2n-1，k+1) R(2n-1，k+1) | G(2n-1，k+2) B(2n-1，k+2) |
| G(2n，k) B(2n，k) | G(2n，k+1) R(2n，k+1) | G(2n，k+2) B(2n，k+2) |
| G(2n+1，k) B(2n+1，k) | G(2n+1，k+1) R(2n+1，k+1) | G(2n+1，k+2) B(2n+1，k+2) |
Table 4A
Table 4B
| G(2n-1，k) R(2n-1，k) | G(2n-1，k+1) R(2n-1，k+1) | G(2n-1，k+2) R(2n-1，k+2) |
| G(2n，k) B(2n，k) | G(2n，k+1) B(2n，k+1) | G(2n，k+2) B(2n，k+2) |
| G(2n+1，k) R(2n+1，k) | G(2n+1，k+1) R(2n+1，k+1) | G(2n+1，k+2) R(2n+1，k+2) |
Claims (58)
1, a kind of image data compressor circuit is in order to export compression image data to a main frame; It is characterized in that, comprising:
One optoelectronic induction module comprises at least two pixel sensors, and in order to pick up corresponding pixel data on the image data, each pixel data of this image data resolves into one first color data, one second color data and one the 3rd color data;
The image data through overcompression in order to receive a control signal and image data and it is compressed processing, is exported then in one signal conversion processes unit;
Wherein, the 2n-1 row of compression image data comprise first color data and part second color data, and the 2n row of compression image data comprise first color data and part the 3rd color data.
2, compressor circuit according to claim 1 is characterized in that, described optoelectronic induction module is a charge coupled cell.
3, compressor circuit according to claim 1 is characterized in that, described optoelectronic induction module is a contact video inductor.
4, compressor circuit according to claim 1 is characterized in that, described signal conversion processes unit comprises:
One analog/digital signal conversion device receives and changes the output signal of described optoelectronic induction module;
One digital signal processing unit receives the output signal of described analog/digital signal conversion device and the control signal of described control module.
5, compressor circuit according to claim 1 is characterized in that, more comprises:
The one scan control module is to provide control signal;
One exports buffer cell, connects the signal output part of described signal conversion processes unit;
One interface circuit is connected between output buffer cell and described main frame.
6, compressor circuit according to claim 1 is characterized in that, described main frame is a PC.
7, compressor circuit according to claim 1 is characterized in that, described first color data is represented green, and described second color data is represented blueness, and the 3rd color data is represented redness.
8, compressor circuit according to claim 1 is characterized in that, described first color data is represented green, and second color data represents that redness and the 3rd color data represent blueness.
9, compressor circuit according to claim 1 is characterized in that, described first color data is represented yellow, and second color data represents that blue-green and the 3rd color data represent aubergine.
10, compressor circuit according to claim 1 is characterized in that, described first color data is represented yellow, and second color data represents that aubergine and the 3rd color data represent blue-green.
11, compressor circuit according to claim 1 is characterized in that, second color data of described compression image data part is to be reference position with first pixel, and the compartment of terrain is disposed in the 2n-1 row.
12, compressor circuit according to claim 1 is characterized in that, second color data of described compression image data part is a reference position with second pixel, and the compartment of terrain is disposed in the 2n-1 row.
13, compressor circuit according to claim 1 is characterized in that, the 3rd color data of described compression image data part is to be reference position with first pixel, and the compartment of terrain is disposed in the 2n row.
14, compressor circuit according to claim 1 is characterized in that, the 3rd color data of described compression image data part is to be reference position with second pixel, and the compartment of terrain is disposed in the 2n row.
15, a kind of image data compressor circuit in order to export compression image data to a main frame, comprising:
One optoelectronic induction module comprises at least two pixel sensors, is used to pick up the corresponding pixel data of image data, and each pixel data of this image data can resolve into one first color data, one second color data and one the 3rd color data;
One signal conversion processes unit receives image data; And
One compressing control unit connects the signal output part of signal conversion processes unit, and exports compression image data signal;
Wherein, the 2n-1 row of compression image data comprise first color data and part second color data, and the 2n row of compression image data comprise first color data and part the 3rd color data.
16, compressor circuit according to claim 15 is characterized in that, described optoelectronic induction module is a charge coupled cell.
17, compressor circuit according to claim 15 is characterized in that, described signal conversion processes unit comprises:
One analog/digital signal conversion device is to receive and to change the output signal of described optoelectronic induction module; And
One digital signal processing unit is with the output signal that receives described analog/digital signal conversion device and the control signal of described control module.
18, compressor circuit according to claim 15 also comprises:
One exports buffer cell, receives the output signal of described signal conversion processes unit; And
The circuit of one interface is coupled between output buffer cell and described main frame.
19, compressor circuit according to claim 15 is characterized in that, described main frame is a PC.
20, compressor circuit according to claim 15 is characterized in that, described first color data is represented green, and second color data represents that blueness and the 3rd color data represent redness.
21, video compressor circuit according to claim 15 is characterized in that, described first color data is represented green, and second color data represents that redness and the 3rd color data represent blueness.
22, compressor circuit according to claim 15 is characterized in that, described first color data is represented yellow, and second color data represents that aubergine and the 3rd color data represent blue-green.
23, video compressor circuit according to claim 15 is characterized in that, described first color data is represented yellow, and second color data represents that blue-green and the 3rd color data represent aubergine.
24, compressor circuit according to claim 15 is characterized in that, second color data of described compression image data part is to be reference position with first pixel, and the compartment of terrain is disposed in the 2n-1 row.
25, compressor circuit according to claim 15 is characterized in that, part second color data of described compression image data is to be reference position with second pixel, and the compartment of terrain is disposed in the 2n-1 row.
26, compressor circuit according to claim 15 is characterized in that, part the 3rd color data of described compression image data is to be reference position with first pixel, and the compartment of terrain is disposed in the 2n row.
27, compressor circuit according to claim 15 is characterized in that, part the 3rd color data of described compression image data is a reference position with second pixel, and the compartment of terrain is disposed in the 2n row.
28, a kind of decompression method of image data, can have capable pixel value P (x of M row * N to one, y) the image data that is compressed decompresses, definition x is row coordinate (x=1～M), y for the row coordinate (y=1～N), this known pixel and color data thereof that is compressed that image data comprises is: (a) row coordinate x, and row coordinate y is when being arbitrary value, (x y) comprises one first color data C1 to pixel value P
Old(x, y); (b) row coordinate x is an odd number, and row coordinate y is when being odd number, and (x y) more comprises one second color data C2 to pixel value P
Old(x, y); And (c) row coordinate x is an even number, and row coordinate y is when being even number, and (x y) more comprises one the 3rd color data C3 to pixel value P
Old(x, y), this video decompression method can (x y) produces the second color data C2 that is short of at first pixel value P
New(x, y) and/or the 3rd color data C3
New(x, y), decompression method may further comprise the steps:
If row coordinate x is odd number, and row coordinate y is when being even number, pixel value P (x, second color data C2 y)
New(x is y) by C1
Old(x, y), C2
Old(x, y-1), C1
Old(x, y-1), C2
Old(x, y-1), C1
Old(x, y+1) institute's decision jointly;
If row coordinate x is even number, and row coordinate y is when being odd number, pixel value P (x, second color data y), C2
New(x is y) by C1
Old(x, y), C2
Old(x-1, y), C1
Old(x-1, y), C2
Old(x+1, y), C1
Old(x+1, y) institute's decision jointly;
If row coordinate x is odd number, and row coordinate y is when being even number, pixel value P (x, the 3rd color data C3 y)
New(x is y) by C1
Old(x, y), C3
Old(x-1, y), C1
Old(x-1, y), C3
Old(x+1, y), C1
Old(x+1, y) institute's decision jointly;
If row coordinate x is even number, and row coordinate y is when being even number, pixel value P (x, second color data C2 y)
New(x is y) by C1
Old(x, y), C1
Old(x-1, y-1), C1
Old(x+1, y-1), C1
Old(x-1, y+1), C1
Old(x+1, y+1), C2
Old(x-1, y-1), C2
Old(x+1, y-1), C1
Old(x-1, y+1), C2
Old(x+1, y+1) institute's decision jointly;
If row coordinate x is odd number, and row coordinate y is when being odd number, pixel value P (x, the 3rd color data C3 y)
New(x is y) by C1
Old(x, y), C1
Old(x-1, y-1), C1
Old(x+1, y-1), C1
Old(x-1, y+1), C1
Old(x+1, y+1), C3
Old(x-1, y-1), C3
Old(x+1, y-1), C3
Old(x-1, y+1), C3
Old(x+1, y+1) institute's decision jointly.
29, a kind of decompression method of image data, can have capable pixel value P (x of M row * N to one, y) the image data that is compressed decompresses, definition x is row coordinate (x=1～M), y for the row coordinate (y=1～N) is compressed known pixel and the color data thereof that image data comprises and is: (a) row coordinate x, when row coordinate y is arbitrary value, (x y) comprises one first color data C1 to pixel value P
Old(x, y); (b) row coordinate x is an odd number, and when row coordinate y was odd number, (x y) more comprised one second color data C2 to pixel value P
Old(x, y); And (c) row coordinate x is an even number, and when row coordinate y was odd number, (x y) more comprised one the 3rd color data C3 to pixel value P
Old(x, y), this video decompression method can (x y) produces the second color data C2 that is short of at first pixel value P
New(x, y) and/or the 3rd color data C3
New(x, y), decompression method may further comprise the steps:
If row coordinate x is an odd number, when row coordinate y is even number, pixel value P (x, second color data C2 y)
New(x is y) by C1
Old(x, y), C2
Old(x, y-1), C1
Old(x, y-1), C2
Old(x, y+1), C1
Old(x, y+1) institute's decision jointly;
If row coordinate x is an even number, when row coordinate y is even number, pixel value P (x, the 3rd color data C3 y)
New(x is y) by C1
Old(x, y), C3
Old(x, y-1), C1
Old(x, y-1), C3
Old(x, y+1), C1
Old(x, y+1) institute's decision jointly;
If row coordinate x is an even number, when row coordinate y is odd number, pixel value P (x, second color data C2 y)
New(x is y) by C1
Old(x, y), C2
Old(x-1, y), C1
Old(x-1, y), C2
Old(x+1, y), C1
Old(x+1, y) institute's decision jointly;
If row coordinate x is an odd number, when row coordinate y is odd number, pixel value P (x, the 3rd color data C3 y)
New(x is y) by C1
Old(x, y), C3
Old(x-1, y), C1
Old(x-1, y), C3
Old(x+1, y), C1
Old(x+1, y) institute's decision jointly;
If row coordinate x is even number, and row coordinate y is when being even number, pixel value P (x, second color data C2 y)
New(x is y) by C1
Old(x, y), C1
Old(x-1, y-1), C1
Old(x+1, y-1), C1
Old(x-1, y+1), C1
Old(x+1, y+1), C2
Old(x-1, y-1), C2
Old(x+1, y-1), C2
Old(x-1, y+1), C2
Old(x+1, y+1) institute's decision jointly.
If row coordinate x is odd number, and row coordinate y is when being even number, pixel value P (x, the 3rd color data C3 y)
New(x is y) by C1
Old(x, y), C1
Old(x-1, y-1), C1
Old(x+1, y-1), C1
Old(x-1, y+1), C1
Old(x+1, y+1) C3
Old(x-1, y-1), C3
Old(x+1, y-1), C3
Old(x-1, y+1), C3
Old(x+1 y+1) is total to the common decision of institute.
30, a kind of decompression method of image data, can have capable pixel value P (x of M row * N to one, y) the image data that is compressed decompresses, definition x is row coordinate (x=1～M), y is row coordinate (y=1～N), being compressed known pixel and the color data thereof that image data comprises is: (a) row coordinate x, when row coordinate y is arbitrary value, (x y) comprises one first color data C1 to pixel value P
Old(x, y); (b) row coordinate x is an odd number, and when row coordinate y was even number, (x y) more comprised one second color data C2 to pixel value P
Old(x, y); And (c) row coordinate x is an even number, and when row coordinate y was odd number, (x y) more comprised one the 3rd color data C3 to pixel value P
Old(x, y), the video decompression method can (x y) produces the second color data C2 that is short of at each pixel value P
New(x, y) and/or the 3rd color data C3
New(x, y), decompression method may further comprise the steps:
If row coordinate x is an odd number, when row coordinate y is odd number, pixel value P (x, second color data C2 y)
New(x is y) by C1
Old(x, y), C2
Old(x, y-1), C1
Old(x, y-1), C2
Old(x, y+1), C1
Old(x, y+1) institute's decision jointly;
If row coordinate x is an even number, when row coordinate y is even number, pixel value P (x, the 3rd color data C3 y)
New(x is y) by C1
Old(x, y), C3
Old(x, y-1), C1
Old(x, y-1), C3
Old(x, y+1), C1
Old(x, y+1) institute's decision jointly;
If row coordinate x is an even number, when row coordinate y is even number, pixel value P (x, second color data C2 y)
New(x is y) by C1
Old(x, y), C2
Old(x-1, y), C1
Old(x-1, y), C2
Old(x+1, y), C1
Old(x+1, y) institute's decision jointly;
If row coordinate x is an odd number, when row coordinate y is odd number, pixel value P (x, the 3rd color data C3 y)
New(x is y) by C1
Old(x, y), C3
Old(x-1, y), C1
Old(x-1, y), C3
Old(x+1, y), C1
Old(x+1, y) institute's decision jointly;
If row coordinate x is an even number, when row coordinate y is odd number, pixel value P (x, second color data C2 y)
New(x is y) by C1
Old(x, y), C1
Old(x-1, y-1), C1
Old(x+1, y-1), C1
Old(x-1, y+1), C1
Old(x+1, y+1), C2
Old(x-1, y-1), C2
Old(x+1, y-1), C2
Old(x-1, y+1), C2
Old(x+1, y+1) institute's decision jointly;
If row coordinate x is an odd number, when row coordinate y is even number, pixel value P (x, the 3rd color data C3 y)
New(x is y) by C1
Old(x, y), C1
Old(x-1, y-1), C1
Old(x+1, y-1), C1
Old(x-1, y+1), C1
Old(x+1, y+1), C3
Old(x-1, y-1), C3
Old(x+1, y-1), C3
Old(x-1, y+1), C3
Old(x+1, y+1) institute's decision jointly.
31, a kind of decompression method of image data, can have capable pixel value P (x of M row * N to one, y) the image data that is compressed decompresses, definition x is row coordinate (x=1～M), y for the row coordinate (y=1～N) is compressed the color data that image data comprises and is: (a) row coordinate x, when row coordinate y is arbitrary value, (x y) comprises one first color data C1 to pixel value P
Old(x, y); (b) row coordinate x is an odd number, and when row coordinate y was even number, (x y) more comprised one second color data C2 to pixel value P
Old(x, y); And (c) row coordinate x is an even number, and row coordinate y is when being even number, and (x y) more comprises one the 3rd color data C3 to pixel value P
Old(x, y), the video decompression method can (x y) produces the second color data C2 that is short of at each pixel value P
New(x, y) and/or the 3rd color data C3
New(x, y), decompression method may further comprise the steps:
If row coordinate x is an odd number, when row coordinate y is odd number, pixel value P (x, second color data C2 y)
New(x is y) by C1
Old(x, y), C2
Old(x, y-1), C1
Old(x, y-1), C2
Old(x, y+1), C1
Old(x, y+1) institute's decision jointly;
If row coordinate x is an even number, when row coordinate y is odd number, pixel value P (x, the 3rd color data C3 y)
New(x is y) by C1
Old(x, y), C3
Old(x, y-1), C1
Old(x, y-1), C3
Old(x, y+1), C1
Old(x, y+1) institute's decision jointly;
If row coordinate x is an even number, when row coordinate y is even number, pixel value P (x, second color data C2 y)
New(x is y) by C1
Old(x, y), C2
Old(x-1, y), C1
Old(x-1, y), C2
Old(x+1, y), C1
Old(x+1, y) institute's decision jointly;
If row coordinate x is an odd number, when row coordinate y is even number, pixel value P (x, the 3rd color data C3 y)
New(x is y) by C1
Old(x, y), C3
Old(x-1, y), C1
Old(x-1, y), C3
Old(x+1, y), C1
Old(x+1, y) institute's decision jointly;
If row coordinate x is an even number, when row coordinate y is odd number, pixel value P (x, second color data C2 y)
New(x is y) by C1
Old(x, y), C1
Old(x-1, y-1), C1
Old(x+1, y-1), C1
Old(x-1, y+1), C1
Old(x+1, y+1), C2
Old(x-1, y-1), C2
Old(x+1, y-1), C2
Old(x-1, y+1), C2
Old(x+1, y+1) institute's decision jointly; And
If row coordinate x is an odd number, when row coordinate y is odd number, pixel value P (x, the 3rd color data C3 y)
New(x is y) by C1
Old(x, y), C1
Old(x-1, y-1), C1
Old(x+1, y-1), C1
Old(x-1, y+1), C1
Old(x+1, y+1), C3
Old(x-1, y-1), C3
Old(x+1, y-1), C3
Old(x-1, y+1), C3
Old(x+1, y+1) institute's decision jointly;
32, a kind of compression method of image data, can be compressed an image data with capable pixel of M row * N, and corresponding to each the pixel P among the image data (x, y), the color constituent of each pixel all comprises one first color data C1 (x, y), one second color data C2 (x, y) and one the 3rd color data C3 (x, y), at each pixel P (x, y), the compression method of image data is:
If the row coordinate x of this pixel is an odd number, when row coordinate y is odd number, export this pixel the first color data C1 (x, y) and the second color data C2 of this pixel (x, y);
If this pixel column coordinate x is odd number, when row coordinate y is even number, export this pixel the first color data C1 (x, y);
If this pixel column coordinate x is even number, when row coordinate y is odd number, export this pixel the first color data C1 (x, y);
If this pixel column coordinate x is even number, when row coordinate y is even number, export this pixel the first color data C1 (x, y) and the 3rd color data C3 of this pixel (x, y).
33, compression method according to claim 32 is characterized in that, also comprises the steps:
If this pixel column coordinate x is odd number, when row coordinate y is even number, also export this pixel the 3rd color data C3 (x, y); And
If this pixel column coordinate x is even number, when row coordinate y is odd number, also export this pixel the second color data C2 (x, y).
34, a kind of compression method of image data, can be compressed an image data with capable pixel of M row * N, corresponding to each the pixel P in this image data (x, y), the color constituent of each pixel all comprises one first color data C1 (x, y), one second color data C2 (x, y) and one the 3rd color data C3 (x, y), at each pixel P (x, y), this compression method may further comprise the steps:
If the row coordinate x of this pixel is an odd number, when row coordinate y is odd number, export this pixel the first color data C1 (x, y);
If the row coordinate x of this pixel is an odd number, when row coordinate y is even number, export this pixel the first color data C1 (x, y) and the second color data C2 of this pixel (x, y);
If the row coordinate x of this pixel is an even number, when row coordinate y is odd number, export this pixel the first color data C1 (x, y) and the 3rd color data C3 of this pixel (x, y);
If the row coordinate x of this pixel is an even number, when row coordinate y is even number, export this pixel the first color data C1 (x, y).
35, compression method according to claim 34 is characterized in that, also comprises the steps:
If this pixel column coordinate x is odd number, when row coordinate y is odd number, also export this pixel the 3rd color data C3 (x, y);
If this pixel column coordinate x is even number, when row coordinate y is even number, also export this pixel the second color data C2 (x, y).
36, a kind of video compression method, can be compressed an image data with capable pixel of M row * N, corresponding to each the pixel P in this image data (x, y), the color constituent of each pixel all comprises one first color data C1 (x, y), one second color data C2 (x, y) and one the 3rd color data C3 (x, y), at each pixel P (x, y), the method for this compression may further comprise the steps:
If the row coordinate x of this pixel is an odd number, when row coordinate y is even number, export this pixel the first color data C1 (x, y);
If the row coordinate x of this pixel is an odd number, when row coordinate y is odd number, export this pixel the first color data C1 (x, y) and the second color data C2 (x, y);
If the row coordinate x of this pixel is an even number, when row coordinate y is odd number, export this pixel the first color data C1 (x, y) and the 3rd color data C3 (x, y);
If the row coordinate x of this pixel is an even number, when row coordinate y is even number, export this pixel the first color data C1 (x, y).
37, compression method according to claim 36 is characterized in that, and is further comprising the steps of:
If the row coordinate x of this pixel is an odd number, when row coordinate y is odd number, also export this pixel the 3rd color data C3 (x, y);
If the row coordinate x of this pixel is an even number, when row coordinate y is odd number, also export this pixel the second color data C2 (x, y).
38, a kind of compression method of image data, can be compressed an image data with capable pixel of M row * N, and corresponding to each the pixel P in this image data (x, y), the color constituent of each pixel all comprises one first color data C1 (x, y), one second color data C2 (x, y) and one the 3rd color data C3 (x, y), at each pixel P (x, y), this compression method may further comprise the steps:
If the row coordinate x of this pixel is an odd number, when row coordinate y is odd number, export this pixel the first color data C1 (x, y);
If the row coordinate x of this pixel is an odd number, when row coordinate y is even number, export this pixel the first color data C1 (x, y) and the second color data C2 of this pixel (x, y);
If this pixel column coordinate x is even number, when row coordinate y is odd number, export this pixel the first color data C1 (x, y) and the 3rd color data C3 of this pixel (x, y);
If the row coordinate x of this pixel is an even number, when row coordinate y is even number, export this pixel the first color data C1 (x, y).
39, according to the described compression method of claim 38, it is characterized in that, further comprising the steps of:
If the row coordinate x of this pixel is an odd number, when row coordinate y is even number, also export this pixel the 3rd color data C3 (x, y);
If the row coordinate x of this pixel is an even number, when row coordinate y is even number, also export this pixel the second color data C2 (x, y).
40, a kind of compression method of image data, can be compressed arranging the image data that forms by a row pixel, and the color constituent corresponding to each pixel P (x) in this image data (coordinate of x for arranging) all comprises one first color data C1 (x) and one second color data C2 (x), and this compression method may further comprise the steps:
When x is even number, export the first color data C1 (x) and the second color data C2 (x) of this pixel;
When x is odd number, export the first color data C1 (x) of this pixel.
According to the described compression method of claim 40, it is characterized in that 41, the color constituent of described each pixel P (x) also comprises one the 3rd color data C3 (x).
According to the described compression method of claim 41, it is characterized in that 42, described step 1 when x is even number, is exported the first color data C1 (x), the second color data C2 (x) and the 3rd color data C3 (x) of this pixel.
According to the described compression method of claim 41, it is characterized in that 43, described step 2 when x is odd number, is exported the first color data C1 (x) and the 3rd color data C3 (x) of this pixel.
44, a kind of compression method of image data, compressed arranging the image data that forms by a row pixel, color constituent corresponding to each the pixel P (x) in this image data, each pixel all comprises one first color data C1 (x) and one second color data C2 (x), and this video compression method may further comprise the steps:
When x is odd number, export first color data C1 (x) of this pixel and the second color data C2 (x) of this pixel;
When x is even number, export the first color data C1 (x) of this pixel.
According to the described compression method of claim 44, it is characterized in that 45, the color constituent of each pixel P (x) also comprises one the 3rd color data C3 (x).
According to the described compression method of claim 45, it is characterized in that 46, described step 1 when x is odd number, is exported the first color data C1 (x), the second color data C2 (x) and the 3rd color data C3 (x) of this pixel.
According to the described compression method of claim 45, it is characterized in that 47, described step 2 when x is even number, is exported the first color data C1 (x) and the 3rd color data C3 (x) of this pixel.
48, a kind of decompression method of image data, to arrange by M pixel P of row (x) form be compressed the image data decompression, x=1～M, this is compressed image data and comprises: (a) coordinate x is by 1 to M, and all pixel P (x) comprise one first color data C1
Old(x), and (b) coordinate x is by 1 to M and when being even number, and pixel P (x) comprises one second color data C2
Old(x), when this video decompression method can produce coordinate x and is odd number, the second color data C2 that pixel P (x) is short of
New(x), decompression method comprises:
Default two weighting parameters W
1, W
2, and W
1+ W
2=1; And
When if x is odd number, the second color data C2 of pixel P (x)
New(x)=C1
Old(x)+W
1* [C2
Old(x-1)-C1
Old(x-1)]+W
2* [C2
Old(x+1)-C1
Old(x+1)].
49, according to the described decompression method of claim 48, it is characterized in that: the described image data that is compressed also comprises: coordinate x is by 1 to M and when the even number, one the 3rd color data C3 of pixel P (x)
Old(x).
50, according to the described decompression method of claim 49, it is characterized in that:
Also be preset with two weighting parameters W
3, W
4, W
3+ W
4=1;
When if x is odd number, one the 3rd color data C3 of pixel P (x)
New(x)=C1
Old(x)+W
3* [C3
Old(x-1)-C1
Old(x-1)]+W
4* [C3
Old(x+1)-C1
Old(x+1)].
51, according to the described decompression method of claim 50, it is characterized in that: the described image data that is compressed also comprises: coordinate x is by 1 to M and when the odd number, one the 3rd color data C3 of pixel P (x)
Old(x).
52, according to the described decompression method of claim 51, it is characterized in that:
Also be preset with two weighting parameters W
3, W
4, W
3+ W
4In fact=1; And
When if x is even number, one the 3rd color data C3 of pixel P (x)
New(x)=C1
Old(x)+W
3* [C3
Old(x-1)-C1
Old(x-1)]+W
4* [C3
Old(x+1)-C1
Old(x+1)].
53, a kind of video decompression method, can to arrange by M pixel P of row (x) form be compressed the image data decompression, x=1～M, this is compressed image data and comprises: (a) coordinate x is by 1 to M, the one first color data C1 of all pixel P (x)
Old(x); (b) coordinate x is by 1 to M and when being odd number, the one second color data C2 of pixel P (x)
Old(x); When this video decompression method can produce coordinate x and is even number, the first color data C2 that pixel P (x) is short of
New(x), this video decompression method may further comprise the steps:
Default two weighting parameters W
1, W
2, W
1+ W
2=1;
When if x is even number, the second color data C2 of pixel P (x)
New(x)=C1
Old(x)+W
1* [C2
Old(x-1)-C1
Old(x-1)]+W
2* [C2
Old(x+1)-C1
Old(x+1)].
According to the described decompression method of claim 53, it is characterized in that 54, the described image data that is compressed also comprises: coordinate x is by 1 to M and when the even number, one the 3rd color data C3 of pixel P (x)
Old(x).
55, according to the described decompression method of claim 54, it is characterized in that:
Also default two weighting parameters W
3, W
4, W
3+ W
4=1;
When if x is odd number, the 3rd color data C3 of pixel P (x)
New(x)=C1
Old(x)+W
3* [C3
Old(x-1)-C1
Old(x-1)]+W
4* [C3
Old(x+1)-C1
Old(x+1)].
According to the described decompression method of claim 53, it is characterized in that 56, the described image data that is compressed also comprises: coordinate x is by 1 to M and when the odd number, one the 3rd color data C3 of pixel P (x)
Old(x).
57, according to the described decompression method of claim 56, it is characterized in that:
Also default two weighting parameters W
3, W
4, W
3+ W
4=1;
When if x is even number, the 3rd color data C3 of pixel P (x)
New(x)=C1
Old(x)+W
3* [C3
Old(x-1)-C1
Old(x-1)]+W
4* [C3
Old(x+1)-C1
Old(x+1)].
58, a kind of compression method of image data, can be compressed an image data with capable the pixel of M row * N, and corresponding to each the pixel P in this image data (x, y), the color constituent of each pixel all comprise one first color data C1 (x, y).One second color data C2 (x, y) and one the 3rd color data C3 (x, y) at each pixel P (x, y), this video compression method may further comprise the steps:
When if the row coordinate x of this pixel is odd number, export this pixel the first color data C1 (x, y) with the second color data C2 (x, y);
When if the row coordinate x of this pixel is even number, export this pixel the first color data C1 (x, y) with and the 3rd color data C3 (x, y).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB991097580A CN1178124C (en) | 1999-07-12 | 1999-07-12 | Image data compressing circuit and compressing and decompressing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB991097580A CN1178124C (en) | 1999-07-12 | 1999-07-12 | Image data compressing circuit and compressing and decompressing method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1280330A CN1280330A (en) | 2001-01-17 |
| CN1178124C true CN1178124C (en) | 2004-12-01 |
Family
ID=5274134
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB991097580A Expired - Fee Related CN1178124C (en) | 1999-07-12 | 1999-07-12 | Image data compressing circuit and compressing and decompressing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1178124C (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7365888B2 (en) * | 2002-07-19 | 2008-04-29 | Brother Kogyo Kabushiki Kaisha | Image read apparatus |
| CN111510154B (en) * | 2020-04-17 | 2023-07-21 | 环荣电子(惠州)有限公司 | Coordinate data compression method |
-
1999
- 1999-07-12 CN CNB991097580A patent/CN1178124C/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| CN1280330A (en) | 2001-01-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1652563A (en) | Image scanning device and its control method | |
| US8009338B2 (en) | Image processing apparatus | |
| JP2009100026A (en) | Image processor | |
| JP4893791B2 (en) | Image processing method and image processing apparatus | |
| EP1954023A2 (en) | Image processing apparatus which makes stored image data suitable for use both internally and in an external apparatus | |
| JP4315209B2 (en) | Image processing apparatus and ground color detection method | |
| CN101064766A (en) | Image processing method and image processing apparatus | |
| CN1842129A (en) | Image processing apparatus and its method | |
| CN1092006C (en) | Image processing apparatus and method | |
| JP4280404B2 (en) | Image processing apparatus and method | |
| CN1515110A (en) | Image encoding method and image device | |
| CN1302149A (en) | Image communication equipment and method | |
| CN1905608A (en) | Image input apparatus and image forming method | |
| CN1178124C (en) | Image data compressing circuit and compressing and decompressing method | |
| US20100245936A1 (en) | Image processing apparatus, image processing method, and program | |
| CN1794777A (en) | Image processing device | |
| CN1941834A (en) | Method and apparatus for image processing | |
| JP2008172677A (en) | Image processor and ground color detection method | |
| CN1604614A (en) | Image processing apparatus | |
| JP2004128664A (en) | Image processor and processing method | |
| JP2728208B2 (en) | Image processing method | |
| CN1825895A (en) | Halftoning device, image processing apparatus, printing apparatus, halftoning method, and program | |
| CN100502457C (en) | Method for filtering color information | |
| JP4781041B2 (en) | Image processing device | |
| US6628438B1 (en) | Method for compressing image data with fixed rate and device therefor |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C06 | Publication | ||
| PB01 | Publication | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| ASS | Succession or assignment of patent right |
Owner name: CHINA NATIONAL HEAVY DUTY TRUCK GROUP JINAN POWER Free format text: FORMER OWNER: JINAN TECH CENTER CO., LTD., CHINA HEAVY TRUCK GROUP Effective date: 20101027 |
|
| C56 | Change in the name or address of the patentee | ||
| CP03 | Change of name, title or address |
Address after: China Taiwan Taoyuan County Patentee after: Qisda Corporation Address before: Taiwan, China Patentee before: Benq Corp. |
|
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20041201 Termination date: 20120712 |