CN101409766A - Image-scaling method and apparatus - Google Patents
Image-scaling method and apparatus Download PDFInfo
- Publication number
- CN101409766A CN101409766A CNA2008101469459A CN200810146945A CN101409766A CN 101409766 A CN101409766 A CN 101409766A CN A2008101469459 A CNA2008101469459 A CN A2008101469459A CN 200810146945 A CN200810146945 A CN 200810146945A CN 101409766 A CN101409766 A CN 101409766A
- Authority
- CN
- China
- Prior art keywords
- horizontal
- vertical
- interpolation
- scaled data
- data
- 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.)
- Pending
Links
Images
Abstract
The embodiment of the invention discloses an image scaling method and a device thereof. The method comprises the following steps: the horizontal cubic interpolation scaling of image data is carried out so as to obtain horizontal image scaling data; and the vertical interpolation of the horizontal image scaling data is carried out so as to obtain full image scaling data. In the embodiment of the invention, the image quality after the scaling is enhanced by applying the cubic interpolation scaling technique.
Description
Technical field
The present invention relates to moving communicating field, particularly relate to a kind of image-scaling method and device.
Background technology
At present increasing mobile phone is supported camera function, digital zoom is absolutely necessary for a multiple functional shooting mobile phone, and the image zoom technology is to realize an of paramount importance ring of hand-set digit zoom function, its principle is based on original pixel in the image exactly, according to certain operation rule under the constraints that as far as possible keeps original primary image information, existing pixel number certificate is handled, increase or reduce with realization pixel number, thereby realize the process that image amplifies or dwindles.The quality of image scaling device scaler (device for zooming) performance directly influences the quality of image and the performance of entire chip system.
A kind of any pantograph ratio realtime graphic device for zooming of the prior art mainly is made up of two parts of horizontal interpolation convergent-divergent and vertical interpolation convergent-divergent.The convergent-divergent process is: the capable pixel data with original image is cached in the horizontal scaling image data buffer earlier, carry out horizontal interpolation by horizontal interpolation device then, capable pixel data behind the convergent-divergent deposits in the vertically scale image data buffer, carry out the interpolation of vertical direction then by vertical interpolation device, thereby finish the convergent-divergent of whole sub-picture.In the convergent-divergent process, adopt the bilinear interpolation algorithm, promptly adopt one 2 * 2 neighborhood, carry out linear interpolation in the horizontal direction earlier, carry out linear interpolation then in vertical direction, thereby obtain new interpolation point.
In realizing process of the present invention, the inventor finds to exist in the prior art following shortcoming:
Because what adopt is the bilinear interpolation algorithm, each new pixel all is to be derived by 4 associated pixel points in the source images (on horizontal direction, the vertical direction each 2) in the purpose image, associated pixel point or insufficient, therefore, cause scaled images contrast and brightness not enough, it is fuzzy that image shows slightly, and image detail quality and expressive force lack perfection.
Summary of the invention
The embodiment of the invention provides a kind of image-scaling method and device thereof, by the application of cubic interpolation zoom technology, has promoted the scaled images quality.
The embodiment of the invention provides a kind of image-scaling method, comprising:
View data is carried out horizontal cubic interpolation convergent-divergent obtain the horizontal image scaled data, described horizontal image scaled data is carried out vertical interpolation convergent-divergent obtain all images scaled data; Or
View data is carried out horizontal interpolation convergent-divergent obtain the horizontal image scaled data, described horizontal image scaled data is carried out vertical cubic interpolation convergent-divergent obtain all images scaled data; Or
View data is carried out vertical cubic interpolation convergent-divergent obtain the vertical image scaled data, described vertical image scaled data is carried out horizontal interpolation convergent-divergent obtain all images scaled data; Or
View data is carried out vertical interpolation convergent-divergent obtain the vertical image scaled data, described vertical image scaled data is carried out horizontal cubic interpolation convergent-divergent obtain all images scaled data.
A kind of image scaling device comprises:
Horizontal interpolating unit is used for that view data is carried out horizontal cubic interpolation convergent-divergent and obtains the horizontal image scaled data;
Vertical interpolating unit is used for that the horizontal image scaled data that described horizontal interpolating unit obtains is carried out vertical interpolation convergent-divergent and obtains all images scaled data.
A kind of image scaling device comprises:
Horizontal interpolating unit is used for that view data is carried out horizontal interpolation convergent-divergent and obtains the horizontal image scaled data;
Vertical interpolating unit is used for that the horizontal image scaled data that described horizontal interpolating unit obtains is carried out vertical cubic interpolation convergent-divergent and obtains all images scaled data.
A kind of image scaling device comprises:
Vertical interpolating unit is used for that view data is carried out vertical cubic interpolation convergent-divergent and obtains the vertical image scaled data;
Horizontal interpolating unit is used for that the vertical image scaled data that described vertical interpolating unit obtains is carried out horizontal interpolation convergent-divergent and obtains all images scaled data.
A kind of image scaling device comprises:
Vertical interpolating unit is used for that view data is carried out vertical interpolation convergent-divergent and obtains the vertical image scaled data;
Horizontal interpolating unit is used for that the vertical image scaled data that described vertical interpolating unit obtains is carried out horizontal cubic interpolation convergent-divergent and obtains all images scaled data.
In the embodiments of the invention,, promoted picture quality behind the convergent-divergent by the application of cubic interpolation zoom technology.
Description of drawings
In order to be illustrated more clearly in the embodiment of the invention or technical scheme of the prior art, to do to introduce simply to the accompanying drawing of required use in embodiment or the description of the Prior Art below, apparently, accompanying drawing in describing below only is some embodiments of the present invention, for those of ordinary skills, under the prerequisite of not paying creative work, can also obtain other accompanying drawing according to these accompanying drawings.
Fig. 1 is a kind of image-scaling method flow chart in the embodiment of the invention;
Fig. 2 is a kind of image-scaling method flow chart in the embodiment of the invention;
Fig. 3 is an interpolation method schematic diagram on a kind of horizontal direction in the embodiment of the invention;
Fig. 4 is that a kind of horizontal location of interpolation is sought and interpolation coefficient computational methods flow chart in the embodiment of the invention;
Fig. 5 is a kind of vertical location of interpolation finding method flow chart in the embodiment of the invention;
Fig. 6 is a kind of image scaling device structure chart in the embodiment of the invention;
Fig. 7 is a kind of image scaling device structure chart in the embodiment of the invention;
Fig. 8 is a kind of image scaling device general structure block diagram in the embodiment of the invention.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the invention, the technical scheme in the embodiment of the invention is clearly and completely described, obviously, described embodiment only is the present invention's part embodiment, rather than whole embodiment.Based on the embodiment among the present invention, those of ordinary skills belong to the scope of protection of the invention not making the every other embodiment that is obtained under the creative work prerequisite.
The embodiment of the invention provides a kind of image-scaling method, and described method may further comprise the steps as shown in Figure 1:
In the embodiment of the invention, provide a kind of image-scaling method,, promoted image quality in images behind the convergent-divergent, positive meaning has been arranged for the popularization of image zoom technology by the application of cubic interpolation zoom technology.
The embodiment of the invention provides a kind of image-scaling method, and described method may further comprise the steps as shown in Figure 2:
Suppose with length to be that the source pixel row of M become the purpose pixel column that length is N, at first, set up two pointers, one is pointed to original pixel column as source pointer m, reads source pixel, and another is as purpose pointer n, the pixel column after the sensing conversion.Now find the insertion position of purpose pixel in the pixel column of source earlier.
With the plain column number of source image is S
0, S
1, S
2..., S
M-2, S
M-1, the purpose pixel column is numbered D
0, D
1, D
2..., D
N-2, D
N-1, purpose pixel D then
nThe insertion position can obtain by following formula:
In formula (1), L represents insertion position and the plain S of source image
0Distance.If the integer of m for obtaining after L is rounded downwards, promptly
Expression rounds down.
Interpolation method as shown in Figure 3 on the horizontal direction.L represents the distance of purpose pixel Dn and source pixel Sm, and then the distance of Dn and Sm-1, Sm+1, Sm+2 is respectively 1+l, 1-l, 2-l.
The interpolation coefficient of cubic interpolation and purpose pixel insertion position are relevant apart from l with source pixel, and interpolation coefficient is a cubic function about distance l, generally adopts the formula of formula (2).In formula (2), p (t) represents interpolation coefficient, and t represents the distance of purpose pixel insertion position with source pixel.
If p1, p2, p3, p4 are respectively the interpolation coefficient of Sm-1, Sm, Sm+1, Sm+2, then
p1＝4-8(1+l)+5(1+l)
2-(1+l)
3＝-l+2l
2-l
3
p2＝1-2l
2+l
3
p3＝1-2(1-l)
2+(1-l)
3＝l+l
2-l
3
p4＝4-8(2-l)+5(2-l)
2-(2-l)
3＝-l
2+l
3
The pixel value of Dn is obtained through interpolation by Sm-1, Sm, Sm+1, Sm+2:
Vn＝Wm-1*p1+Wm*p2+Wm+1*p3+Wm+2*p4
＝W
m-1*(-l+2l
2-l
3)+W
m*(1-2l
2+l
3)+W
m+1*(l+l
2-l
3)+W
m+2*(-l
2+l
3)
(3)
In the horizontal direction, suppose that the wide of image is scaled N by M.By formula (1) as can be seen, the L as a result that obtains at last is a floating number.Can obtain by further analysis, purpose pixel of every insertion, L just increases (M-1)/(N-1).Therefore, if can make high-precision quantification, just can eliminate floating-point operation to (M-1)/(N-1).
In the multi mode terminal chip of using, in order to make error, be 21 bits to the fractional part quantisation depth of (M-1)/(N-1) in reasonable range and so on.Maximum minification when dwindling according to image again can obtain the integer part of (M-1)/(N-1).Such as if maximum minification is 4, promptly the purpose image narrows down to 1/4 o'clock of source images, and integer part (M-1)/(N-1) is greater than 4, and the integer greater than 4 will need 3 bits with binary representation, and therefore, the total quantization degree of depth is the 21+3=24 bit.Then in hardware was realized, l represented to get final product with 24 bits.Specifically quantization method is as the formula (4):
Omit the fractional part of L1, and what only keep its integer part hangs down 24, just obtain horizontal interpolation constant C h, if the calculating of this constant can be used complicated resources such as divider with the hardware realization, the present invention has calculated the value of Ch earlier by software, passes to device for zooming scaler by register configuration mode, and horizontal interpolation controller only need simply be added and subtracted calculating according to the value of Ch, calculate interpolation coefficient h, and seek location of interpolation according to above-mentioned algorithm.
Horizontal interpolation calculation of the present invention is mutual by software and hardware, the mode of decomposition interpolation coefficient computational process realizes, the value of Ch is realized by software, interpolation coefficient is calculated by hardware, in the implementation procedure of reality, the decomposition that interpolation coefficient calculates can be different, and parameter can all be drawn by computed in software, or realize by hardware that all perhaps software and hardware comes mutual by other parameter (non-Ch).
The flow chart that horizontal location of interpolation is sought and interpolation coefficient h calculates may further comprise the steps as shown in Figure 4:
Step 401 is with source pixel pointer m, purpose pixel pointer n, horizontal interpolation coefficient h initialize 0;
Step 404 judges that whether h is less than 2
21, if then change step 405, otherwise change step 407;
Step 406 judges whether n equals N-1, if, then finish, if not, step 402 then changeed;
Realized that by this method the position of current interpolating pixel seeks, and calculated the interpolation coefficient of current pixel easily, and offered horizontal interpolation calculation device and carry out the calculating of pixel value.
During horizontal interpolation coefficient calculates,, may take different quantified precisions and quantification means eliminating the floating number stage.Such as, can allow the quantified precision of fractional part of interpolation constant be 22 bits, at this moment,
Again such as, may produce the interpolation constant according to the methods below:
Step 202 is carried out horizontal cubic interpolation according to described horizontal cubic interpolation position and horizontal cubic interpolation coefficient and is calculated the horizontal image scaled data.
The A interpolation calculation is optimized
If carry out interpolation arithmetic, need 7 multipliers at least according to formula (3).Can obtain by analyzing:
Vn＝W
m-1*(-l+2l
2-l
3)+W
m*(1-2l
2+l
3)+W
m+1*(l+l
2-l
3)+W
m+2*(-l
2+l
3)
＝l{l[(W
m+2+W
m-W
m-1-W
m+1)l+(2W
m-1-2W
m+W
m+1-W
m+2)]+(W
m+1-W
m-1)}+W
m
Make A=W
M+2+ W
m-W
M-1-W
M+1,
B＝2W
m-1-2W
m+W
m+1-W
m+2，
C＝W
m+1-W
m-1，
Then:
V
n＝l*{l*[A*l+B]+C}+W
m (5)
As can be seen, after optimizing, the realization of interpolation calculation reduces to multiplication 3 times from 7 multiplication.
B multiplier cut position
As can be seen, A is one 10 a signed number, and B is one 11 a signed number, and C is one 9 a signed number, and l is one 24 a unsigned number.If cut position is not carried out in the multiplying of formula (5), promptly computational accuracy is made a choice, bit wide can be increasing.Therefore, carrying out cut position at needs handles.
Adopt such error analysis method to carry out cut position: the purpose image that obtains according to floating-point operation with the purpose figure image subtraction of selecting to obtain after the computational accuracy, calculate the mean value and the variance of this difference then, select computational accuracy according to the mean value and the variance of difference.
According to experimental result repeatedly, take following method to carry out cut position in the horizontal direction and handle.
A) in all multiplyings, l is got 8, be specially [20:13];
B) A*l be 10 take advantage of 8 18 multiplication;
C) A*l+B＜＜8 be 18 add 18 19 additions, high 16 to 19 reservations that obtain;
D) l* (A*l+B＜＜8) be 8 take advantage of 16 24 multiplication, high 16 to 24 reservations that obtain;
E) l* (A*l+B＜＜8)+C＜＜5 be 16 add 14 17 additions, high 16 to 17 reservations that obtain;
F) l*[l* (A*l+B＜＜8)+C＜＜5] be 8 take advantage of 16 24 multiplication, 24 of obtaining are kept 9, be specially [20:12];
G) l*[l* (A*l+B＜＜8)+C＜＜5]+Wm be 9 add 8 10 additions, 10 of obtaining are kept least-significant bytes;
Horizontal interpolation calculation device of the present invention is taked the method for suitable cut position on the basis of statistical error analysis, but also can realize the function of this cubic interpolation with other cut position mode, carry out multiplying such as l being got 9, after carrying out multiplying, keep the position of different numbers.
2) bit width conversion and buffering storage
Processing for the ease of vertically scale, improve interpolation efficiency, this part can be converted to 32bit by 8bit with the data through horizontal scaling, promptly store 4 pixel datas among a word, vertical interpolation then uses line buffer to carry out buffer-stored the data after the bit width conversion all based on the word operation afterwards.
Step 203 is carried out vertical interpolation convergent-divergent to described horizontal image scaled data and is obtained all images scaled data.
The present invention can select for use linear interpolation algorithm to realize vertically scale in vertical direction, but the present invention can select for use the cubic interpolation algorithm to realize the arbitrary proportion convergent-divergent of vertical direction epigraph equally.
If select for use linear interpolation algorithm to realize vertically scale in vertical direction, then hypothesis is high is scaled T by R, takes approach based on linear interpolation to seek location of interpolation.
Be similar to horizontal interpolation algorithm, omit the fractional part of L1, keep its integer part, just obtained vertical interpolation constant C v.The method of vertical interpolation may further comprise the steps with flow chart as shown in Figure 5:
Step 501 is with source pixel pointer r, purpose pixel pointer t, vertical interpolation coefficient v initialize 0;
Step 504 judges that whether v is less than 2
21, if then change step 505, otherwise change step 507;
Step 506 judges whether t equals T-1, if, then finish, if not, step 502 then changeed;
Simultaneously, signal according to the mac input is differentiated the bus As-Is, if bus is congested, vertical interpolation controller can suspend in line buffer peek interpolation, can utilize the storage resources among the line buffer that data are cushioned, when bus is idle a little, restart vertical interpolation operation, the crowded of bus avoided in data output, improved image zoom efficient.
The pixel value of Dt can be obtained through interpolation by Sr and Sr+1:
Vt＝Wr*(1-l)+Wr+1*l＝Wr+(Wr+1-Wr)*l (7)
The embodiment of the invention provides a kind of image-scaling method, at the prior art zooming effect is not very desirable, on the basis that error statistics is analyzed, a kind of 4 cubic interpolations that carry out have in the horizontal direction been proposed, carry out the image-scaling method of 2 linear interpolation in vertical direction.The present invention carries out interpolation (because a point on the vertical direction is obtained by 4 point interpolations on the horizontal direction according to 8 associated pixel points in the original image, so carry out each pixel of 2 images after the linear interpolation in vertical direction is the equal of to obtain according to 8 associated pixel points in the original image), the image that obtains all promotes to some extent than the contrast and the brightness of the image that prior art obtains; Simultaneously, owing to taked to eliminate floating-point operation and take suitable effective measures such as cut position method, reduced the cost of cubic interpolation on hardware is realized.The present invention is ingenious in addition decomposes algorithm, and the calculating of partial parameters is finished by software, passes to hardware then, finishes whole zoom operations.
Through the above description of the embodiments, those skilled in the art can be well understood to the present invention and can realize by hardware, also can realize based on such understanding by the mode that software adds necessary general hardware platform, technical scheme of the present invention can embody with the form of software product, it (can be CD-ROM that this software product can be stored in a non-volatile memory medium, USB flash disk, portable hard drive etc.) in, comprise that some instructions are with so that a computer equipment (can be a personal computer, server, the perhaps network equipment etc.) carry out the described method of each embodiment of the present invention.
The embodiment of the invention provides a kind of image scaling device, as shown in Figure 6, comprising: horizontal interpolating unit 601 is used for that view data is carried out horizontal cubic interpolation convergent-divergent and obtains the horizontal image scaled data; Vertical interpolating unit 602 is used for that the horizontal image scaled data that described horizontal interpolating unit 601 obtains is carried out vertical interpolation convergent-divergent and obtains all images scaled data.
In the embodiment of the invention, provide a kind of image scaling device,, promoted image quality in images behind the convergent-divergent, positive meaning has been arranged for the popularization of image zoom technology by the application of cubic interpolation zoom technology.
The embodiment of the invention provides a kind of image scaling device, and on the basis of above-mentioned image scaling device shown in Figure 6, as shown in Figure 7, comprising: horizontal interpolating unit 601 is used for that view data is carried out horizontal cubic interpolation convergent-divergent and obtains the horizontal image scaled data; Bit width conversion unit 603 is used for the horizontal image scaled data that described horizontal interpolating unit 601 obtains is carried out bit width conversion, and the bit wide of described horizontal image scaled data is converted to the second bit wide value by the first bit wide value; Data buffer storage unit 604 is used for the horizontal image scaled data that described bit width conversion unit 603 obtains is carried out buffer-stored; Vertical interpolating unit 602 is used for that the horizontal image scaled data of described vertical data buffer cell 604 buffer memorys is carried out vertical interpolation convergent-divergent and obtains all images scaled data.
Wherein, described horizontal interpolating unit 601 comprises: horizontal interpolation control unit 605, the horizontal cubic interpolation coefficient that is used to seek the horizontal cubic interpolation position of described view data and determines described view data; Horizontal interpolation calculation unit 606, the horizontal cubic interpolation position and the horizontal cubic interpolation coefficient that are used for obtaining according to described horizontal interpolation control unit 605 calculate the horizontal image scaled data.
In the embodiment of the invention, provide a kind of image scaling device,, promoted image quality in images behind the convergent-divergent, positive meaning has been arranged for the popularization of image zoom technology by the application of cubic interpolation zoom technology.
In the embodiment of the invention, provide a kind of image scaling device,, promoted image quality in images behind the convergent-divergent, positive meaning has been arranged for the popularization of image zoom technology by the application of cubic interpolation zoom technology.
In the embodiment of the invention, a kind of image scaling device is provided, application by the cubic interpolation zoom technology, promoted image quality in images behind the convergent-divergent, and bit width conversion and metadata cache technology have been adopted, improved image zoom efficient, positive meaning has been arranged for the popularization of image zoom technology.
The embodiment of the invention provides a kind of general structure block diagram of image scaling device, as shown in Figure 8:
The horizontal interpolation part of scaler is by horizontal data buffer (buffer storage), horizontal interpolation controller, and horizontal interpolation calculation device three parts are formed.Scaler is after external interface such as camera interface reception data, Y (brightness) UV (contrast) data are deposited in respectively among separately horizontal data buffer:hbuf_y, hbuf_u, the hbuf_v, each buffer can store the yuv data of 4 8bit bit wides, calculates for cubic interpolation and uses.Horizontal interpolation controller passes through to seek location of interpolation, and calculates every interpolation coefficient, calculates interpolation point by horizontal interpolation device according to adjacent four points of storing among the horizontal data buffer at last, it may be noted that Y data and UV data are separate processes.
Vertical interpolation part is made up of vertical data buffer line buffer, vertical interpolation calculation device, vertical interpolation controller three parts.Vertical data buffer line buffer can storing one row pass through the view data of horizontal scaling, and exports the data of 4 pairs of points in adjacent two row simultaneously.Vertical interpolation controller is exported to the interpolation calculation device by seeking the interpolation coefficient that location of interpolation calculates every row with interpolation coefficient.Vertical interpolation calculation device once calculates 4 points, thereby finishes vertical interpolation operation according to 4 pairs of points in adjacent two row of line buffer input and interpolation coefficient.
In the embodiment of the invention, a kind of image scaling device is provided, application by the cubic interpolation zoom technology, promoted the brightness and contrast of image behind the convergent-divergent, and bit width conversion and metadata cache technology have been adopted, improved image zoom efficient, positive meaning has been arranged for the popularization of image zoom technology.
Below only be specific embodiments more of the present invention, those skilled in the art can also be according to the record of the foregoing description, expects being equal to or the scheme of conversion of other easily, for example, horizontal direction adopts linear interpolation to carry out convergent-divergent, adopts cubic interpolation to carry out convergent-divergent on the vertical direction; Or horizontal direction and vertical direction all adopt cubic interpolation to carry out convergent-divergent; Or carry out the convergent-divergent of vertical direction earlier, carry out the convergent-divergent of horizontal direction again.Be equal to or the scheme of conversion according to above-mentioned,, can adapt to the method for record or device and be equal to or conversion, obtain other execution mode in conjunction with the write up of the embodiment of front.Therefore protection scope of the present invention should comprise that these significantly are equal to or the scheme of conversion etc.For example, the image-scaling method of the embodiment of the invention can also be for as follows: view data is carried out horizontal interpolation convergent-divergent obtain the horizontal image scaled data, described horizontal image scaled data is carried out vertical cubic interpolation convergent-divergent obtain all images scaled data; Or view data is carried out vertical cubic interpolation convergent-divergent obtain the vertical image scaled data, described vertical image scaled data is carried out horizontal interpolation convergent-divergent obtain all images scaled data; Or view data is carried out vertical interpolation convergent-divergent obtain the vertical image scaled data, described vertical image scaled data is carried out horizontal cubic interpolation convergent-divergent obtain all images scaled data.The scheme of these conversion is different is the direction difference of carrying out cubic interpolation, or it is different to carry out the order of convergent-divergent on horizontal direction and the vertical direction, but the same front of concrete principle and method repeats no more.For image scaling device, then, can obtain accordingly according to the introduction of front, promptly the embodiment of the invention also provides following image scaling device:
A kind of image scaling device comprises:
Horizontal interpolating unit is used for that view data is carried out horizontal interpolation convergent-divergent and obtains the horizontal image scaled data;
Vertical interpolating unit is used for that the horizontal image scaled data that described horizontal interpolating unit obtains is carried out vertical cubic interpolation convergent-divergent and obtains all images scaled data.
Correspondingly, described vertical interpolating unit comprises:
Vertical interpolation control unit is used to seek the vertical cubic interpolation position of described view data and the vertical cubic interpolation coefficient of definite described view data;
Vertical interpolation calculation unit, the vertical cubic interpolation position that is used for obtaining according to described vertical interpolation control unit calculates the horizontal image scaled data with vertical cubic interpolation coefficient.
A kind of image scaling device comprises:
Vertical interpolating unit is used for that view data is carried out vertical cubic interpolation convergent-divergent and obtains the vertical image scaled data;
Horizontal interpolating unit is used for that the vertical image scaled data that described vertical interpolating unit obtains is carried out horizontal interpolation convergent-divergent and obtains all images scaled data.
A kind of image scaling device comprises:
Vertical interpolating unit is used for that view data is carried out vertical interpolation convergent-divergent and obtains the vertical image scaled data;
Horizontal interpolating unit is used for that the vertical image scaled data that described vertical interpolating unit obtains is carried out horizontal cubic interpolation convergent-divergent and obtains all images scaled data.
In a word, the above is preferred embodiment of the present invention only, is not to be used to limit protection scope of the present invention.Within the spirit and principles in the present invention all, any modification of being done, be equal to replacement, improvement etc., all should be included within protection scope of the present invention.
Claims (12)
1, a kind of image-scaling method is characterized in that, may further comprise the steps:
View data is carried out horizontal cubic interpolation convergent-divergent obtain the horizontal image scaled data, described horizontal image scaled data is carried out vertical interpolation convergent-divergent obtain all images scaled data; Or
View data is carried out horizontal interpolation convergent-divergent obtain the horizontal image scaled data, described horizontal image scaled data is carried out vertical cubic interpolation convergent-divergent obtain all images scaled data; Or
View data is carried out vertical cubic interpolation convergent-divergent obtain the vertical image scaled data, described vertical image scaled data is carried out horizontal interpolation convergent-divergent obtain all images scaled data; Or
View data is carried out vertical interpolation convergent-divergent obtain the vertical image scaled data, described vertical image scaled data is carried out horizontal cubic interpolation convergent-divergent obtain all images scaled data.
2, the method for claim 1 is characterized in that, describedly view data is carried out horizontal cubic interpolation convergent-divergent obtains the horizontal image scaled data and comprises:
Seek the horizontal cubic interpolation position of described view data and the horizontal cubic interpolation coefficient of definite described view data;
Calculate the horizontal image scaled data according to described horizontal cubic interpolation position and described horizontal cubic interpolation coefficient.
3, the method for claim 1 is characterized in that, describedly the horizontal image scaled data is carried out vertical interpolation convergent-divergent obtains all images scaled data and comprises:
Described horizontal image scaled data is carried out vertical cubic interpolation convergent-divergent obtain all images scaled data; Or
Described horizontal image scaled data is carried out vertical linearity interpolation convergent-divergent obtain all images scaled data.
4, the method for claim 1 is characterized in that, describedly the horizontal image scaled data is carried out vertical interpolation convergent-divergent obtains all images scaled data and comprises:
Described horizontal image scaled data is carried out bit width conversion, the bit wide of described horizontal image scaled data is converted to the second bit wide value by the first bit wide value;
Described horizontal image scaled data through bit width conversion is carried out buffer-stored.
5, the method for claim 1 is characterized in that, describedly the horizontal image scaled data is carried out vertical interpolation convergent-divergent obtains also comprising before all images scaled data:
Differentiate the bus situation, when bus is congested, suspends and read described horizontal image scaled data; When bus is idle, restarts and read described horizontal image scaled data and follow-up vertical interpolation procedure.
6, a kind of image scaling device is characterized in that, comprising:
Horizontal interpolating unit is used for that view data is carried out horizontal cubic interpolation convergent-divergent and obtains the horizontal image scaled data;
Vertical interpolating unit is used for that the horizontal image scaled data that described horizontal interpolating unit obtains is carried out vertical interpolation convergent-divergent and obtains all images scaled data.
7, as image scaling device as described in the claim 6, it is characterized in that, also comprise:
The bit width conversion unit is used for described horizontal image scaled data is carried out bit width conversion, and the bit wide of described horizontal image scaled data is converted to the second bit wide value by the first bit wide value;
Data buffer storage unit is used for the horizontal image scaled data that described bit width conversion unit obtains is carried out buffer-stored.
8, as image scaling device as described in the claim 6, it is characterized in that described horizontal interpolating unit comprises:
Horizontal interpolation control unit, the horizontal cubic interpolation coefficient that is used to seek the horizontal cubic interpolation position of described view data and determines described view data;
Horizontal interpolation calculation unit, the horizontal cubic interpolation position and the horizontal cubic interpolation coefficient that are used for obtaining according to described horizontal interpolation control unit calculate the horizontal image scaled data.
9, a kind of image scaling device is characterized in that, comprising:
Horizontal interpolating unit is used for that view data is carried out horizontal interpolation convergent-divergent and obtains the horizontal image scaled data;
Vertical interpolating unit is used for that the horizontal image scaled data that described horizontal interpolating unit obtains is carried out vertical cubic interpolation convergent-divergent and obtains all images scaled data.
10, as image scaling device as described in the claim 9, it is characterized in that described vertical interpolating unit comprises:
Vertical interpolation control unit is used to seek the vertical cubic interpolation position of described view data and the vertical cubic interpolation coefficient of definite described view data;
Vertical interpolation calculation unit, the vertical cubic interpolation position that is used for obtaining according to described vertical interpolation control unit calculates the horizontal image scaled data with vertical cubic interpolation coefficient.
11, a kind of image scaling device is characterized in that, comprising:
Vertical interpolating unit is used for that view data is carried out vertical cubic interpolation convergent-divergent and obtains the vertical image scaled data;
Horizontal interpolating unit is used for that the vertical image scaled data that described vertical interpolating unit obtains is carried out horizontal interpolation convergent-divergent and obtains all images scaled data.
12, a kind of image scaling device is characterized in that, comprising:
Vertical interpolating unit is used for that view data is carried out vertical interpolation convergent-divergent and obtains the vertical image scaled data;
Horizontal interpolating unit is used for that the vertical image scaled data that described vertical interpolating unit obtains is carried out horizontal cubic interpolation convergent-divergent and obtains all images scaled data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2008101469459A CN101409766A (en) | 2007-12-27 | 2008-08-27 | Image-scaling method and apparatus |
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200710198676.6 | 2007-12-27 | ||
| CN200710198676 | 2007-12-27 | ||
| CNA2008101469459A CN101409766A (en) | 2007-12-27 | 2008-08-27 | Image-scaling method and apparatus |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101409766A true CN101409766A (en) | 2009-04-15 |
Family
ID=40572537
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2008101469459A Pending CN101409766A (en) | 2007-12-27 | 2008-08-27 | Image-scaling method and apparatus |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101409766A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010142128A1 (en) * | 2009-06-12 | 2010-12-16 | 深圳市融创天下科技发展有限公司 | Method for adaptively scaling image |
| CN101616249B (en) * | 2009-07-28 | 2011-08-10 | 谭洪舟 | Video scaling device based on bicubic interpolation |
| CN102333212A (en) * | 2010-07-14 | 2012-01-25 | 北京大学 | Bilinear two-fold upsampling method and system thereof |
| CN103034973A (en) * | 2012-12-05 | 2013-04-10 | 焦点科技股份有限公司 | Self-adaptive image scaling method based on bicubic interpolation |
| CN103544678A (en) * | 2012-07-13 | 2014-01-29 | 浙江大华技术股份有限公司 | Video image processing device and video image processing method |
| CN108009983A (en) * | 2017-11-30 | 2018-05-08 | 天津天地基业科技有限公司 | A kind of method that artwork is corresponded to based on graphical analysis positioning pixel |
| WO2019041264A1 (en) * | 2017-08-31 | 2019-03-07 | 深圳市大疆创新科技有限公司 | Image processing apparatus and method, and related circuit |
| CN110363723A (en) * | 2019-07-16 | 2019-10-22 | 安健科技(广东)有限公司 | Improve the image processing method and device of image boundary effect |
| CN110992383A (en) * | 2020-03-03 | 2020-04-10 | 天津精诊医疗科技有限公司 | CT image liver artery segmentation method and system based on deep learning |
| CN114092338A (en) * | 2022-01-20 | 2022-02-25 | 长沙金维信息技术有限公司 | Image zooming fast calculation method |
-
2008
- 2008-08-27 CN CNA2008101469459A patent/CN101409766A/en active Pending
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010142128A1 (en) * | 2009-06-12 | 2010-12-16 | 深圳市融创天下科技发展有限公司 | Method for adaptively scaling image |
| CN101616249B (en) * | 2009-07-28 | 2011-08-10 | 谭洪舟 | Video scaling device based on bicubic interpolation |
| CN102333212A (en) * | 2010-07-14 | 2012-01-25 | 北京大学 | Bilinear two-fold upsampling method and system thereof |
| CN102333212B (en) * | 2010-07-14 | 2013-07-17 | 北京大学 | Bilinear two-fold upsampling method and system thereof |
| CN103544678A (en) * | 2012-07-13 | 2014-01-29 | 浙江大华技术股份有限公司 | Video image processing device and video image processing method |
| CN103034973B (en) * | 2012-12-05 | 2015-08-12 | 焦点科技股份有限公司 | Based on the adaptive image scaling method of bicubic interpolation |
| CN103034973A (en) * | 2012-12-05 | 2013-04-10 | 焦点科技股份有限公司 | Self-adaptive image scaling method based on bicubic interpolation |
| WO2019041264A1 (en) * | 2017-08-31 | 2019-03-07 | 深圳市大疆创新科技有限公司 | Image processing apparatus and method, and related circuit |
| CN108009983A (en) * | 2017-11-30 | 2018-05-08 | 天津天地基业科技有限公司 | A kind of method that artwork is corresponded to based on graphical analysis positioning pixel |
| CN108009983B (en) * | 2017-11-30 | 2021-05-25 | 天地伟业技术有限公司 | Method for positioning pixel point corresponding to original image based on image analysis |
| CN110363723A (en) * | 2019-07-16 | 2019-10-22 | 安健科技(广东)有限公司 | Improve the image processing method and device of image boundary effect |
| CN110363723B (en) * | 2019-07-16 | 2021-06-29 | 安健科技(广东)有限公司 | Image processing method and device for improving image boundary effect |
| CN110992383A (en) * | 2020-03-03 | 2020-04-10 | 天津精诊医疗科技有限公司 | CT image liver artery segmentation method and system based on deep learning |
| CN114092338A (en) * | 2022-01-20 | 2022-02-25 | 长沙金维信息技术有限公司 | Image zooming fast calculation method |
| CN114092338B (en) * | 2022-01-20 | 2022-04-19 | 长沙金维信息技术有限公司 | Image zooming fast calculation method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101409766A (en) | Image-scaling method and apparatus | |
| KR100310584B1 (en) | A system for signal processing using multiply-add operations | |
| CN101572771B (en) | Device, system, and method for solving systems of linear equations using parallel processing | |
| CN110519603B (en) | Hardware circuit for real-time video zooming and zooming method thereof | |
| JPH09326958A (en) | Image processing unit and processing method | |
| CN108492338B (en) | Compression method and device for animation file, storage medium and electronic device | |
| CN108363559A (en) | Multiplication processing method, equipment and the computer-readable medium of neural network | |
| CN1326395C (en) | Data processing method,data processing device | |
| KR930001689A (en) | Electronic zooming system using image buffer | |
| CN112565890B (en) | Video clipping method and device, storage medium and electronic equipment | |
| WO2016186782A1 (en) | Video encoding and decoding | |
| CN116306709A (en) | Data processing method, medium and electronic equipment | |
| US20230010981A1 (en) | Methods and apparatuses for high performance and accuracy fixed-point scale implementation | |
| US20210082082A1 (en) | Data processing method and processing circuit | |
| CN114201140A (en) | Exponential function processing unit, method and neural network chip | |
| JP2000187726A (en) | Data interpolation method and its device and storage medium | |
| JPH0744354A (en) | Signal processor | |
| JP4920559B2 (en) | Data processing device | |
| US8135069B2 (en) | Frame rate conversion method using motion interpolation | |
| JP5072558B2 (en) | Data processing device | |
| CN108335706B (en) | Seamless playing method of multimedia file, terminal equipment and storage medium | |
| US7191203B2 (en) | Method and system for high-speed multiplication | |
| US8407274B2 (en) | Machine division | |
| WO2014030257A1 (en) | Processing device and processing method | |
| CN113438051A (en) | Encoding method, encoding device, and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Open date: 20090415 |