CN102170518B - System and method for realizing non-linear zoom of video image - Google Patents

System and method for realizing non-linear zoom of video image Download PDF

Info

Publication number
CN102170518B
CN102170518B CN201010116296.5A CN201010116296A CN102170518B CN 102170518 B CN102170518 B CN 102170518B CN 201010116296 A CN201010116296 A CN 201010116296A CN 102170518 B CN102170518 B CN 102170518B
Authority
CN
China
Prior art keywords
linear
length
interpolation
cosine function
zoom
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201010116296.5A
Other languages
Chinese (zh)
Other versions
CN102170518A (en
Inventor
李琛
林晓伟
严卫健
石岭
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenzhen Shenyang electronic Limited by Share Ltd
Original Assignee
Arkmicro Technologies Inc
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Arkmicro Technologies Inc filed Critical Arkmicro Technologies Inc
Priority to CN201010116296.5A priority Critical patent/CN102170518B/en
Publication of CN102170518A publication Critical patent/CN102170518A/en
Application granted granted Critical
Publication of CN102170518B publication Critical patent/CN102170518B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Landscapes

  • Studio Circuits (AREA)

Abstract

The invention discloses a system for realizing non-linear zoom of a video image. The system is characterized by comprising a cosine function generator, a step length calculation module, a step length accumulator and an interpolator. In the invention, a cosine function amplitude is input into the cosine function generator to obtain an output value, the output value and linear step length are input into the step length calculation module to obtain non-linear step length, the non-linear step length is inputted into the step length accumulator to obtain position information of an interpolation point, the position information and an original pixel point are input into the interpolator, and an interpolation is output according to a selected interpolation algorithm to obtain a pixel point value. The invention also discloses a method for realizing non-linear zoom of a video image based on the system for realizing non-linear zoom of a video image. By utilizing the system and method for realizing the non-linear zoom of the video image, the features of a cosine curve are utilized, and a non-linear zoomed curve is fitted, so as to well realize the non-linear zoom of the video image and the smooth transition of each zoomed area.

Description

A kind of system and method realizing video image non-linear zoom
Technical field
The present invention relates to a kind of video image processing system and method, relate to a kind of method and system of video image being carried out to convergent-divergent process especially.
Background technology
Image scaling is a kind of technology conventional in video or image display system, because the resolution of display device is different with the resolution of input original image, therefore needs to zoom in or out to original image to adapt to output display unit.When image length and width by equal proportion convergent-divergent time, effect is best, such as the Nonlinear magnify of 640 × 480 resolution to 1024 × 768, image length and width equal proportion amplifies 1.6 times.But usually there is the situation that the Aspect Ratio of display device is different from input picture Aspect Ratio, such as motion picture screen and many large-size screen monitors TVs adopt the display format of 16:9 at present, and traditional computer or television image are the form of 4:3, when on the display device that the image of 4:3 will be put into 16:9 during display all over the screen, due to can not length and width equal proportion convergent-divergent, just there will be image scene by the sensation of flattening up and down, impact viewing comfort level, as shown in Figure 1a.
The method general expression of current this problem of solution mends black surround or horizontal nonlinearity convergent-divergent to image level.Mend black surround effect as shown in Figure 1 b, image can not be all over the screen; Non-linear zoom is relatively applicable to display all over the screen, as illustrated in figure 1 c.Larger magnification ratio is adopted in the marginal portion of image, and adopt smaller magnification ratio at the middle section of image, because the main body of image is generally positioned at mid portion, time therefore with non-linear zoom viewing image, be unlikely to have too large discomfort.
Contrary, the image of 16:9 is presented on the display device of 4:3, also needs to use non-linear scaling techniques.
Image is divided into 3 regions in existing technical scheme, is centre/left/right region respectively, adopts different conic sections to control the scaling of image, as Fig. 2 to each region.For the Nonlinear magnify of 4:3 to 16:9, if input image resolution is H_in × V_in, display device resolution is H_out × V_out, then horizontal direction needs amplify multiple be:
H_ratio=H_out/H_in (1)
Suppose that former figure zone line length is c_len, this region multiplication factor is c_ratio, then c_ratio should be less than H_ratio, and equally, suppose that the length in former about figure region is s_leh, multiplication factor is that s_ratio, s_ratio should be greater than H_ratio.The magnification ratio of entire image is a conic section, makes to be seamlessly transit from centre to both sides.Though the method can reach the effect of non-linear zoom, realize more complicated, and in block and block junction, need the seamless link of conic section energy, be not easy to control.
Summary of the invention
The technical problem to be solved in the present invention is the non-linear zoom realizing image, realizes seamlessly transitting of image regional convergent-divergent.For solving the problems of the technologies described above, the present invention proposes a kind of cosine function that adopts to realize the system of non-linear zoom, this system comprises cosine function generator, step size computation module, step-length accumulator, interpolation device, wherein input cosine function amplitude value to cosine function generator, the output obtained inputs to step calculation unit together with linear step-length, obtain non-linear step-length, non-linear step-length inputs to step-length accumulator, obtain interpolation point positional information and input to interpolation device, preimage vegetarian refreshments also inputs to interpolation device, according to the pixel point value that the interpolation algorithm output interpolation selected obtains.
The interpolation algorithm selected in described interpolation device can be all with any one in the interpolation algorithm of step-length determination interpolation point.
The invention allows for a kind of method of the non-linear zoom adopting described non-linear zoom system to realize, the method comprises the steps:
Step S401: according to linear step-length and the cosine amplitude of input, generate cosine function by cosine function generator;
Step S402: in step calculation unit, is superimposed to linear step-length by the cosine function obtained, and obtains non-linear step-length;
Step S403: in step-length accumulator, adds up non-linear step-length, obtains the position of the corresponding original image vegetarian refreshments of each interpolation point;
Step S404: obtain interpolation coefficient according to interpolation algorithm and interpolation point positional information and participate in the original image vegetarian refreshments of interpolation, carrying out interpolation calculation, obtains pixel value corresponding to interpolation point and exports.
The cosine function produced in described step S401 is:
C(m)=Acos(2π·m/M),
Wherein, A is cosine amplitude, and M is that output is counted, and m is current interpolation point, and the span of m is the natural number of 0 to M-1.
The computing formula of described non-linear step-length is:
NLP=Acos(2π·m/M)+LP,
Wherein, NLP is non-linear step-length, and LP is linear step-length.
The computational methods of described linear step-length LP are when adopting linear scale, the step-length between interpolation point.
Described cosine amplitude A is by register configuration, and configurable magnitude range is: negative linear step grows to positive linear step-length.
In described step S403, the method that specifically obtains of each interpolation point is that non-linear step-length often accumulates once, and obtains an interpolation point.
Interpolation algorithm described in step S404 can be all with any one in the interpolation algorithm of step-length determination interpolation point.
Beneficial effect of the present invention shows: the present invention utilizes the feature of cosine curve, the curve of fit non-linear convergent-divergent, achieve the non-linear zoom of image well, make the regional transitions smooth of convergent-divergent, and the method realized is simpler, be applicable to various Image Zooming Algorithm, as linear interpolation method, cube interpolation method etc.
Accompanying drawing explanation
Fig. 1 a be 4:3 image when 16:9 display device shows, the schematic diagram of horizontal direction Linear Amplifer;
Fig. 1 b be 4:3 image when 16:9 display device shows, horizontal direction mends the schematic diagram of black surround;
Fig. 1 c be 4:3 image when 16:9 display device shows, the schematic diagram of horizontal direction Image magnify;
Fig. 2 is the schematic diagram of cosine curve one-period;
Fig. 3 is the structured flowchart of video image non-linear zoom system in the specific embodiment of the invention;
Fig. 4 is the flow chart of the non-linear zoom method adopting non-linear zoom system in the specific embodiment of the invention;
Fig. 5 is by the interpolation point position view of non-linear zoom algorithm application when linear interpolation is amplified in the specific embodiment of the invention.
Embodiment
Below in conjunction with accompanying drawing, the specific embodiment of the present invention is described in detail.
Be illustrated in figure 2 characteristic curve C (t) of cosine function, if A is the amplitude of cosine curve, B is the biased of cosine function, i.e. mean value.
C(t)=Acos(2π·t/T)+B (1)
In this embodiment, the characteristic that the non-linear zoom utilizing cosine function and the present invention to need to realize is similar, is superimposed to the step-length of linear scale, obtains the step-length of non-linear zoom, thus realize the non-linear zoom of image by cosine function.Be illustrated in figure 3 the structured flowchart of video image non-linear zoom system described in this embodiment, this panntographic system comprises cosine function generator 301, step size computation module 302, step-length accumulator 303, interpolation device 304, wherein input the cosine function amplitude value A of setting to cosine function generator 301, the output obtained inputs to the non-linear step-length NLP that step calculation unit 302 obtains needed for non-linear zoom together with linear step-length LP, non-linear step-length NLP inputs to step-length accumulator 303, obtain interpolation point positional information and input to interpolation device 304, interpolation device 304 is according to the interpolation point positional information obtained in the interpolation algorithm adopted and step-length accumulator 303, obtain information and the interpolation coefficients such as the number of the preimage vegetarian refreshments participated in required for current point interpolation and reconnaissance, with the original image vegetarian refreshments inputing to interpolation device, export the point that interpolation obtains.
This embodiment is applicable to all convergent-divergent algorithms, as linear interpolation, bi-cubic interpolation etc., to different convergent-divergent algorithms, described cosine function generator 301, step calculation unit 302, step-length accumulator 303 are all identical, interpolation reconnaissance just in interpolation device 304, interpolation coefficient difference, and interpolation computing method is not identical yet.This embodiment is described in detail to non-linear zoom method and system of the present invention for linear interpolation.
The interpolation algorithm of linear interpolation is identical with prior art, and when P point is between x1 point and x2 point, the pixel value that P point is corresponding is be added after pixel value that x1 and x2 is corresponding is multiplied by corresponding interpolation coefficient respectively to obtain.And the pixel participating in interpolation in linear interpolation is two original image vegetarian refreshments about interpolation point position, and the step-length of interpolation is depended in the position of interpolation point.
When linear scale, step-length is determined by scaling; When non-linear zoom, with scaling be below the Nonlinear magnify of 4:3 to 16:9, and be only specifically described for horizontal direction, then the magnification ratio of horizontal direction is:
H_ratio=H_out/H_in (2)
Wherein, H_ratio is horizontal direction magnification ratio, and M is that horizontal direction output is counted, and m is current interpolation point.Then linear step-length is LP=H_in/H_out, then as shown in Figure 4, described in this embodiment, the performing step of non-linear zoom is specific as follows:
Step S401: according to linear step-length and the cosine amplitude of input, generate cosine function by cosine function generator 301, wherein, cosine amplitude is designated as A, and by register configuration, and configurable magnitude range is (-LP, + LP), namely negative linear step grows to positive linear step-length.When cosine amplitude is for time negative, then the scaling of image level direction zone line is larger; Otherwise then the scaling of zone line is little compared with fringe region.As formula (3) cosine function for obtaining:
C(m)=Acos(2π·m/M) (3)
Wherein, the value of m is the natural number of 0 to M-1.
Step S402: in step calculation unit 302, is superimposed to linear step-length LP by the cosine function described in the formula obtained (3), obtains non-linear step-length NLP, as shown in formula (4):
NLP=Acos(2π·m/M)+LP (4)
Step S403: in step-length accumulator 303, adds up non-linear step-length, obtains the position of the corresponding original image vegetarian refreshments of each interpolation point.
Being illustrated in figure 5 input and output counts than a schematic diagram for 6:8 non-linear hour convergent-divergent, and input 6 original image vegetarian refreshments, obtain 8 points after interpolation, step-length often accumulates once, and obtains an interpolation point.Certainly, the present invention is not limited to the scaling shown in this embodiment Fig. 5, and Fig. 5 is only the schematic diagram that a kind of non-linear step-length adds up.
Step S404: obtain interpolation coefficient according to interpolation algorithm and interpolation point positional information and participate in the original image vegetarian refreshments of interpolation, carrying out interpolation calculation, obtains pixel value corresponding to interpolation point and exports.This embodiment is for linear interpolation, then according to the position of the corresponding original image vegetarian refreshments of the interpolation point that obtains in step S403, the point obtaining participating in interpolation is two original pixel values about interpolation point, and interpolation coefficient is also determined by interpolation algorithm; Thus interpolation calculation adopts the pixel point value of participation interpolation calculation and interpolation coefficient weighting to obtain the pixel value of interpolation point.
It should be noted that, non-linear zoom method of the present invention is not limited to linear interpolation, when adopting other interpolation algorithm, after then step-length adds up and obtains interpolation point, then according to different interpolation algorithms in step S404, obtain original image vegetarian refreshments and the interpolation coefficient of different participation interpolation, obtain the pixel value output of interpolation point simultaneously according to different interpolation methods in step s 404.
Above content is in conjunction with concrete preferred implementation further description made for the present invention, can not assert that specific embodiment of the invention is confined to these explanations.For general technical staff of the technical field of the invention, without departing from the inventive concept of the premise, some simple deduction or replace can also be made, all should be considered as belonging to protection scope of the present invention.

Claims (5)

1. one kind realizes the system of video image non-linear zoom, it is characterized in that, this system comprises cosine function generator, step size computation module, step-length accumulator, interpolation device, wherein input cosine function amplitude value to cosine function generator, the output obtained inputs to step calculation unit together with linear step-length, obtain non-linear step-length, non-linear step-length inputs to step-length accumulator, obtain interpolation point positional information and input to interpolation device, preimage vegetarian refreshments also inputs to interpolation device, according to the pixel point value that the interpolation algorithm output interpolation selected obtains;
The cosine function that described cosine function generator obtains is: C (m)=Acos (2 π m/M),
Wherein, A is cosine amplitude, and M is that output is counted, and m is current interpolation point, and the span of m is the natural number of 0 to M-1; Described cosine amplitude A is by register configuration, and configurable magnitude range is: negative linear step grows to positive linear step-length;
The formula that described step calculation unit calculates non-linear step-length is:
NLP=Acos(2π·m/M)+LP,
Wherein, NLP is non-linear step-length, and LP is linear step-length; The computational methods of described linear step-length LP are when adopting linear scale, the step-length between interpolation point.
2. a kind of system realizing video image non-linear zoom according to claim 1, is characterized in that, the interpolation algorithm selected in described interpolation device can be all with any one in the interpolation algorithm of step-length determination interpolation point.
3. adopt the non-linear zoom method that non-linear zoom system as claimed in claim 1 realizes, it is characterized in that, the method comprises the steps:
Step S401: according to linear step-length and the cosine amplitude of input, generate cosine function by cosine function generator, described cosine function is:
C(m)=Acos(2π·m/M),
Wherein, A is cosine amplitude, and M is that output is counted, and m is current interpolation point, and the span of m is the natural number of 0 to M-1; Described cosine amplitude A is by register configuration, and configurable magnitude range is: negative linear step grows to positive linear step-length;
Step S402: in step calculation unit, is superimposed to linear step-length by the cosine function obtained, and obtains non-linear step-length, and the computing formula of described non-linear step-length is:
NLP=Acos(2π·m/M)+LP,
Wherein, NLP is non-linear step-length, and LP is linear step-length; The computational methods of described linear step-length LP are when adopting linear scale, the step-length between interpolation point;
Step S403: in step-length accumulator, adds up non-linear step-length, obtains the position of the corresponding original image vegetarian refreshments of each interpolation point;
Step S404: obtain interpolation coefficient according to interpolation algorithm and interpolation point positional information and participate in the original image vegetarian refreshments of interpolation, carrying out interpolation calculation, obtains pixel value corresponding to interpolation point and exports.
4. non-linear zoom method according to claim 3, is characterized in that, in described step S403, the method that specifically obtains of each interpolation point is that non-linear step-length often accumulates once, and obtains an interpolation point.
5. non-linear zoom method according to claim 3, is characterized in that, interpolation algorithm described in step S404 can be all with any one in the interpolation algorithm of step-length determination interpolation point.
CN201010116296.5A 2010-02-26 2010-02-26 System and method for realizing non-linear zoom of video image Active CN102170518B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201010116296.5A CN102170518B (en) 2010-02-26 2010-02-26 System and method for realizing non-linear zoom of video image

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201010116296.5A CN102170518B (en) 2010-02-26 2010-02-26 System and method for realizing non-linear zoom of video image

Publications (2)

Publication Number Publication Date
CN102170518A CN102170518A (en) 2011-08-31
CN102170518B true CN102170518B (en) 2015-03-04

Family

ID=44491479

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201010116296.5A Active CN102170518B (en) 2010-02-26 2010-02-26 System and method for realizing non-linear zoom of video image

Country Status (1)

Country Link
CN (1) CN102170518B (en)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6650790B1 (en) * 2000-06-09 2003-11-18 Nothshore Laboratories, Inc. Digital processing apparatus for variable image-size enlargement with high-frequency bandwidth synthesis
CN101197957A (en) * 2007-12-25 2008-06-11 上海广电集成电路有限公司 Non-linear image zooming method and system
CN101365077A (en) * 2008-09-28 2009-02-11 深圳华为通信技术有限公司 Image scaling method and device

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050190982A1 (en) * 2003-11-28 2005-09-01 Matsushita Electric Industrial Co., Ltd. Image reducing device and image reducing method

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6650790B1 (en) * 2000-06-09 2003-11-18 Nothshore Laboratories, Inc. Digital processing apparatus for variable image-size enlargement with high-frequency bandwidth synthesis
CN101197957A (en) * 2007-12-25 2008-06-11 上海广电集成电路有限公司 Non-linear image zooming method and system
CN101365077A (en) * 2008-09-28 2009-02-11 深圳华为通信技术有限公司 Image scaling method and device

Also Published As

Publication number Publication date
CN102170518A (en) 2011-08-31

Similar Documents

Publication Publication Date Title
CN103366342B (en) Be applied to the subsection linearity inser value method that video image amplifies
CN103929599A (en) Digital video image real-time zooming method based on FPGA
TWI426775B (en) Camera recalibration system and the method thereof
CN103236035B (en) Based on the image magnification method without the bilateral quadratic B-spline interpolation of skew
WO2010034202A1 (en) Image zooming method and apparatus
CN101572828A (en) Method for correcting distortion in real time based on GPU camera and video camera
TW201120805A (en) Forward and backward resizing method
CN110276739B (en) Video jitter removal method based on deep learning
CN103745429B (en) Method for rapidly realizing eye image processing
CN106780336B (en) Image reduction method and device
CN204948182U (en) A kind of image raising frequency system and display unit
CN102186044A (en) Edge correlation image stepless scaling algorithm and hardware realization device thereof
CN102547067B (en) Improved bicubic interpolation video scaling method
CN100596173C (en) Image zooming method and apparatus
CN103325109A (en) Method used for correcting distortion of fish-eye image and applicable to wall-mounted panoramic video camera
CN101950242A (en) Multiple viewpoint scene imaging scaling display control method
CN102842123B (en) Sparse-region residual error compensating and revising method for improving marginal definition during image sampling
CN102170518B (en) System and method for realizing non-linear zoom of video image
CN104519266A (en) Photographing device and adjusting system and adjusting method thereof
CN102065253B (en) Software and hardware coordinated image horizontal zooming method and device
CN110390636A (en) Unmanned plane is super, the simulation Zooming method of high definition picture or video data
CN103139525B (en) Video image nonlinearity amplification method based on interest region detection
CN101789120B (en) Image interpolation method based on cosine polynomial
CN103327261A (en) Method for simply and fast redirecting video
TWI435286B (en) Image processing method and apparatus

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
CP01 Change in the name or title of a patent holder

Address after: 4, building 406, room 518057, two software park, hi tech Zone, Nanshan District, Guangdong, Shenzhen

Patentee after: Shenzhen Shenyang electronic Limited by Share Ltd

Address before: 4, building 406, room 518057, two software park, hi tech Zone, Nanshan District, Guangdong, Shenzhen

Patentee before: Aike Chuangxin Microelectronic Co., Ltd.

CP01 Change in the name or title of a patent holder