CN1489760A - Method and system for improving display resolution in images using sub-pixel sampling and visual error filtering - Google Patents

Method and system for improving display resolution in images using sub-pixel sampling and visual error filtering Download PDF

Info

Publication number
CN1489760A
CN1489760A CNA018226019A CN01822601A CN1489760A CN 1489760 A CN1489760 A CN 1489760A CN A018226019 A CNA018226019 A CN A018226019A CN 01822601 A CN01822601 A CN 01822601A CN 1489760 A CN1489760 A CN 1489760A
Authority
CN
China
Prior art keywords
image
sps
channel
resolution
sub
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.)
Granted
Application number
CNA018226019A
Other languages
Chinese (zh)
Other versions
CN1267882C (en
Inventor
S・J・戴利
S·J·戴利
K・科武里
R·R·K·科武里
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.)
Sharp Corp
Original Assignee
Sharp Corp
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
Priority claimed from US09/735,425 external-priority patent/US6807319B2/en
Priority claimed from US09/735,454 external-priority patent/US6775420B2/en
Priority claimed from US09/735,424 external-priority patent/US6608632B2/en
Application filed by Sharp Corp filed Critical Sharp Corp
Publication of CN1489760A publication Critical patent/CN1489760A/en
Application granted granted Critical
Publication of CN1267882C publication Critical patent/CN1267882C/en
Anticipated expiration legal-status Critical
Expired - Fee Related legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T3/00—Geometric image transformations in the plane of the image
    • G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
    • G06T3/4015—Image demosaicing, e.g. colour filter arrays [CFA] or Bayer patterns
    • G—PHYSICS
    • G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
    • G09G—ARRANGEMENTS OR CIRCUITS FOR CONTROL OF INDICATING DEVICES USING STATIC MEANS TO PRESENT VARIABLE INFORMATION
    • G09G5/00—Control arrangements or circuits for visual indicators common to cathode-ray tube indicators and other visual indicators
    • G09G5/02—Control arrangements or circuits for visual indicators common to cathode-ray tube indicators and other visual indicators characterised by the way in which colour is displayed
    • G—PHYSICS
    • G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
    • G09G—ARRANGEMENTS OR CIRCUITS FOR CONTROL OF INDICATING DEVICES USING STATIC MEANS TO PRESENT VARIABLE INFORMATION
    • G09G5/00—Control arrangements or circuits for visual indicators common to cathode-ray tube indicators and other visual indicators
    • G09G5/36—Control arrangements or circuits for visual indicators common to cathode-ray tube indicators and other visual indicators characterised by the display of a graphic pattern, e.g. using an all-points-addressable [APA] memory
    • G09G5/39—Control of the bit-mapped memory
    • G09G5/395—Arrangements specially adapted for transferring the contents of the bit-mapped memory to the screen
    • G—PHYSICS
    • G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
    • G09G—ARRANGEMENTS OR CIRCUITS FOR CONTROL OF INDICATING DEVICES USING STATIC MEANS TO PRESENT VARIABLE INFORMATION
    • G09G2320/00—Control of display operating conditions
    • G09G2320/02—Improving the quality of display appearance
    • G—PHYSICS
    • G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
    • G09G—ARRANGEMENTS OR CIRCUITS FOR CONTROL OF INDICATING DEVICES USING STATIC MEANS TO PRESENT VARIABLE INFORMATION
    • G09G2340/00—Aspects of display data processing
    • G09G2340/04—Changes in size, position or resolution of an image
    • G09G2340/0407—Resolution change, inclusive of the use of different resolutions for different screen areas
    • G—PHYSICS
    • G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
    • G09G—ARRANGEMENTS OR CIRCUITS FOR CONTROL OF INDICATING DEVICES USING STATIC MEANS TO PRESENT VARIABLE INFORMATION
    • G09G2340/00—Aspects of display data processing
    • G09G2340/04—Changes in size, position or resolution of an image
    • G09G2340/0407—Resolution change, inclusive of the use of different resolutions for different screen areas
    • G09G2340/0414—Vertical resolution change
    • G—PHYSICS
    • G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
    • G09G—ARRANGEMENTS OR CIRCUITS FOR CONTROL OF INDICATING DEVICES USING STATIC MEANS TO PRESENT VARIABLE INFORMATION
    • G09G2340/00—Aspects of display data processing
    • G09G2340/04—Changes in size, position or resolution of an image
    • G09G2340/0407—Resolution change, inclusive of the use of different resolutions for different screen areas
    • G09G2340/0421—Horizontal resolution change
    • G—PHYSICS
    • G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
    • G09G—ARRANGEMENTS OR CIRCUITS FOR CONTROL OF INDICATING DEVICES USING STATIC MEANS TO PRESENT VARIABLE INFORMATION
    • G09G2340/00—Aspects of display data processing
    • G09G2340/04—Changes in size, position or resolution of an image
    • G09G2340/0407—Resolution change, inclusive of the use of different resolutions for different screen areas
    • G09G2340/0428—Gradation resolution change
    • G—PHYSICS
    • G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
    • G09G—ARRANGEMENTS OR CIRCUITS FOR CONTROL OF INDICATING DEVICES USING STATIC MEANS TO PRESENT VARIABLE INFORMATION
    • G09G2340/00—Aspects of display data processing
    • G09G2340/04—Changes in size, position or resolution of an image
    • G09G2340/0457—Improvement of perceived resolution by subpixel rendering
    • G—PHYSICS
    • G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
    • G09G—ARRANGEMENTS OR CIRCUITS FOR CONTROL OF INDICATING DEVICES USING STATIC MEANS TO PRESENT VARIABLE INFORMATION
    • G09G3/00—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes
    • G09G3/20—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes for presentation of an assembly of a number of characters, e.g. a page, by composing the assembly by combination of individual elements arranged in a matrix no fixed position being assigned to or needed to be assigned to the individual characters or partial characters
    • G09G3/2007—Display of intermediate tones
    • G09G3/2059—Display of intermediate tones using error diffusion
    • G—PHYSICS
    • G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
    • G09G—ARRANGEMENTS OR CIRCUITS FOR CONTROL OF INDICATING DEVICES USING STATIC MEANS TO PRESENT VARIABLE INFORMATION
    • G09G3/00—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes
    • G09G3/20—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes for presentation of an assembly of a number of characters, e.g. a page, by composing the assembly by combination of individual elements arranged in a matrix no fixed position being assigned to or needed to be assigned to the individual characters or partial characters
    • G09G3/2007—Display of intermediate tones
    • G09G3/2074—Display of intermediate tones using sub-pixels

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Hardware Design (AREA)
  • Image Processing (AREA)
  • Control Of Indicators Other Than Cathode Ray Tubes (AREA)
  • Color Image Communication Systems (AREA)
  • Color Television Systems (AREA)
  • Editing Of Facsimile Originals (AREA)
  • Facsimile Image Signal Circuits (AREA)

Abstract

The present invention provides a method and an apparatus for converting a color image of a higher-resolution to an image of a lower-resolution. With the method of the present invention, based on characteristics of spatial frequency responses of the human visual system, a lower-resolution image (116) is formed by separating a higher-resolution image (72) into a luminance channel (76) and chrominance channels (98 & 100) so as to carry out suitable sampling and filtering for each of the luminance channel (76) and the chrominance channels (98 & 100), then by combing the luminance channel (86) and the chrominance channels (111 & 114) after the sampling. As to the chrominance channels (98 & 100), traditional sub-sampling (101 & 103) is carried out to avoid chromatic aliasing, while sub-pixel sampling (80) is performed for the luminance channel (76) to improve resolution of luminance component.

Description

By using sub-pixel sampling and visual error filtering to improve the method and system of the display resolution of image
Technical field
Embodiments of the invention relate in the field with display of high resolution images on the display of low resolution, and wherein display uses three colour cells to dispose the R of display image, G and B or other components.It is general that this three colour cells are configured in the LCD display of direct viewing, and for example, and in such configuration, single pixel is made up of 3 sub-pixels side by side.Each sub-pixel is only controlled primary colours in three primary colours (that is, R, G and B), and the primary colours that each sub-pixel itself is only shown by digital picture are usually controlled.High-resolution image can be available in storer, maybe can be directly available from algorithm (vector graphics, some font design, and computer graphics).
The application's theme is the U.S. Patent application No.09/735 of " Methods and Systems for Improving DisplayResolution using Sub-Pixel Sampling and Visual ErrorCompensation (compensating the method and system that improves display resolution by using sub-pixel sampling and visual error) " with that submit on Dec 12nd, 2000, that invented by Scott Daly, exercise question, and 454 is relevant.This patented claim is quoted at this, for your guidance.
The application's theme is the U.S. Patent application No.09/735 of " Methods and Systems forImproving Display Resolution in Achromatic Images usingSub-Pixel Sampling and Visual Error Compensation (improving the method and system of the display resolution of monochrome image by using sub-pixel sampling and visual error filtering) " with that submit on Dec 12nd, 2000, that invented by Rajesh Reddy KKovvuri and Scott Daly, exercise question, and 425 is relevant.This patented claim is quoted at this, for your guidance.
Background technology
The most normally used method that is used for the image of display of high resolution on the display of low resolution is the resolution that the pixel 2 of high-resolution image 4 is down sampled to lower resolution displays 6, as shown in Figure 1.Then, R, the G of the colour element 8 of each down-sampling, B numerical value are transformed into the R that separates, G, the B element 10,12,14 of each display pixel 16.These R, the G of display pixel 16, B element 10,12,14 are also referred to as sub-pixel.Because display device does not allow color-element overlapping, sub-pixel is only got a kind of color in R, G or three kinds of colors of B, yet the amplitude of color can be in (for example, 0-255) the interior variation of whole tonal range.Sub-pixel have usually 1: 3 the ratio of width to height (width: highly) so that the pixel that obtains at last 16 is square.Do not consider in sampling/converter technique, the R of display, G, the B sub-pixel is a spatial placement; In fact, they are assumed to be with overlapping with their identical modes in high-resolution image as shown in Figure 1.
The pixel of high-definition picture 4 be shown as three a little skew pile up square 8, the RGB numerical value of representing them is to interrelate with same locus (that is pixel).A display pixel 16 that comprises every kind of sub-pixel in R, G, the B sub-pixel 10,12,14, on Fig. 1 by using dark line to be shown as the part of low resolution three colour cell displays 6.Other display pixels are represented with more shallow dotted line.
In this example, high image in different resolution has the resolution bigger 3 times than monitor resolution (on level and vertical dimension).Because this direct sub sampling technology causes the aliasing artefact, has used the whole bag of tricks, average such as adjacent unsampled pixel for the pixel of sampling.Should be pointed out that this when sub sampling the general technology of average adjacent element, on mathematics, be equivalent to the high-resolution image of rectangular filter pre-filtering.In addition, the technology that should be pointed out that the pixel (as shown in this figure) of selecting to be different from the left side can be looked at as the pre-filtering that only influences phase place.Therefore, the great majority relevant with stoping aliasing are handled the filtering operation that can be looked at as for high-definition picture, even only nuclear (Kernel) is applied on the location of pixels of sampling.
Monochrome image as what stipulate, does not have visible colored the variation in this explanation and claim.When this monochromatic condition can occur in image and only comprises one deck or a chrominance channel, or occur in image and have multilayer or a plurality of chrominance channel, but each color layer produces the image of solid color when being identical thus.
See that aforesaid technology is not utilized potential monitor resolution.The background information of this respect can be obtained by the following document of reference: R.Fiegenblatt (1989), " Full color imaging on amplitude color mosaic displays (inlaying full color imaging on the display) " at the amplitude colour, Proc.SPIE V.1075,199-205; And J.Kranz and L.Silverstein (1990); " Color matrix displayimage quality:The effects of luminance and spatialsampling (colour matrix image quality: the effect of brightness and spatial sampling) "; SIDSymp.Digest 29-32; these articles are quoted at this, for your guidance.For example, in demonstration shown in Figure 1, when the resolution of display pixel 16 be high-definition picture (source images) 4 resolution 1/3 the time, sub-pixel 10,12 and 14 resolution equal the resolution (on horizontal scale) in source.Used by the people of colour blindness if this only shows, then might utilize the locus of sub-pixel.This method is shown on following Fig. 2, and wherein the R of Xian Shiing, G and B sub-pixel 10,12 and 14 are to obtain from the corresponding colour of the different pixel 11,13 of high-definition picture and 15.This allows horizontal resolution is subpixel resolution, and it is three times of display picture element resolution.
Is but how for the beholder's (just, most of spectators) who is not achromate's display? for the display slip-stick artist, fortunately, also colour blindness under the highest spatial frequency even have the observer of perfect colored 5 eyesights.This is shown in following Fig. 3, shows the Utopian spatial frequency response of people's vision system on the figure.
Here, brightness 17 is meant the monochromatic visual observation of the image of being watched, and colourity 19 is meant color content, and it is treated to from redness to equal intensification modulation green and from the blueness to the yellow by vision system.The colour difference signal R-G and the B-Y of video image approximately are similar to these modulation.For most of observers, the bandwidth of colour frequency response be the luminance frequency response bandwidth 1/2.Sometimes, the bandwidth of blue-yellow modulated response even littler drops to about 1/3 of brightness.Comprise the sampling of color-element three colour cell sub-pixel conversion, as shown in Figure 2, can be called as sub-pixel sampling from different image pixels to display picture element.
With reference to Fig. 4, on the horizontal direction of display, a frequency range is arranged, be between Nyquist (Nyquist) frequency (each sub-pixel 0.5 cycle=1.5 cycle/threes' colour cell pixel) of Nyquist (Nyquist) frequency (display pixel=three colour cell pixels provides three colour cell Nyquist for each 0.5 cycle of three colour cell pixels) of display pixel 16 and sub-pixel element 10,12 and 14.This zone is shown as rectangular area 20 on Fig. 4.High-definition picture and its width are equaled to show that the rectangular function of sampling interval carries out convolution and the sinc function that draws is shown as shallow dot-and-dash line 22.This is for the method in common that shows that MIF (modulation transfer function) carries out that modeling takes when display is LCD.
The sinc function that draws is shown as dotted line 24 with equaling that sub-pixel rectangle at interval carries out convolution high-resolution source image, and it has higher bandwidth.This is the restriction that is applied by display when considering that sub-pixel is rectangle on a yardstick.In shown rectangular area 20, sub-pixel can display brightness information, but do not have chromatic information.In fact, any chromatic information in this zone all is confused.Therefore, in this zone, by allowing colored aliasing, we can obtain compared with by three colour cells (that is display) frequency monochrome information that pixel allowed, higher.This obtains " favourable " zone by using sub-pixel sampling.
For the application that font shows, the black and white font is typically pretreated, as shown in Figure 5.The standard pre-service comprises hint, and it is meant at the pixel center place and centers for the font stroke, that is, and and the phase shift that the font stroke is specific.This back is followed low-pass filtering usually, is also referred to as the gray scale antialiasing.
Vision frequency response (CSF) shown in Figure 3 is idealized.In fact, they have limited descending slope, shown in Fig. 6 (a).Brightness CSF 30 is transformed into display pixel territory (supposing the viewing distance of 1280 pixels) from the unit of cy/deg (cycles/).It is shown as solid line 30, have to approach for 1.5 cycles/maximum frequency of pixel (display pixel), and it is being that band is logical in shape, has to approach for 0.2 cycle/peak value of pixel three colour cells.R:G CSF32 is shown as dotted line, and low pass just has and approaches for 0.5 cycle/maximum frequency of pixel.B:Y modulation CSF 34 be shown as dot-dash, the LPF curve, have the maximum frequency that is similar to R:G CSF, but have lower peak response.Chrominance C SF 32 and 34 and the cutoff frequency of brightness CSF30 between scope be that wherein we can allow colored aliasing so that improve the zone of brightness bandwidth.
Fig. 6 (a) also shows Utopian image power spectrum 36, as the function of 1/f, is rendered as the straight line (because this figure uses logarithmic axis) with slope of-1 on figure.This frequency spectrum will repeat with sample frequency.These repeat to be shown as the sampling rate for the horizontal direction of pixel 38 and sub-pixel 40.Repeating of occurring of lower frequency 38 is because pixel sampling, and is because sub-pixel sampling repeating of occurring of higher frequency 40.Should be pointed out that alteration of form is because we are drawn in it on the logarithm frequency axis.Expand to frequencies lower frequency, these repetition frequency spectrums that are lower than nyquist frequency and be called as aliasing.Repeating of the left side is that colour obscures 38, because it is because pixel sampling speed, and brightness is obscured 40 and appeared at higher frequency, because it relates to higher sub-pixel sampling speed.
On Fig. 6 (a), do not apply pre-filtering for the source frequency spectrum.Therefore, because the aliasing (that is, colour is obscured) that pixel sampling causes expands to low-down frequency 35.Therefore, even colored CSF has narrower bandwidth compared with brightness CSF, colored artefact still is (contrast that depends on noise and display) that can see.
On Fig. 6 (b), we apply pre-filtering (equaling the rectangular function of three source image pixels) for the source power spectrum, on Fig. 4, be shown as dot-and-dash line 22, and can see that it can influence the baseband frequency spectrum 42 that surpasses 0.5 cycle/pixel, make it have the more precipitous slope in ratio-1, be shown as 44.These repeat also to show the effect of this pre-filtering.Even by this filtering, we see, some colored aliasing (at the frequency spectrum of the repetition at lower frequency place) occurs in frequency 46 places of the cutoff frequency that is lower than two chrominance C SF 32a and 34a.Therefore, can see that simple brightness pre-filtering will be difficult to remove colored aliasing, can not remove all luminance frequency (that is " favourable " zone) that surpass 0.5 cycle/pixel.
Because we depend on the function of the difference of vision system on bandwidth as brightness or colourity, giving us brightness bandwidth in " favourable zone " 20 promotes, so a possibility is according to design pre-filtering as the vision system model of describing in following article: C.Betrisey etc. (2000); " Displaced filtering for patterneddisplays (being used for the displacement filtering of pattern device) "; SID Symposium digest; 296-299; this article is quoted at this; for your guidance, and be shown in Fig. 7.
That color layer and that color sub-pixel that this technology is being sampled according to image are ideally used different prefilters.Like this, 9 wave filters are arranged.They can be designed by the vision difference model that uses the people who describes in following article: X.Zhang and B.Wandell (1996); " A spatial extension of CIELab for digitalcolor image reproduction (spatial spread that is used for the CIELab of digital color image reproduction) "; SID Symp.Digest 731-734; this article is quoted at this; for your guidance, and be shown in Fig. 7.This is that off-line is finished, and supposes image black and white always.In last embodiment, in order to save calculating, the wave filter that uses rectangular function rather than finally obtain.In addition, still old some can the colored mistake of descried residue because colored aliasing expands to the frequency lower than colored CSF cutoff frequency (as what see) downwards on Fig. 6 (b).
Yet employed vision mode is not considered the coverage character of vision system, and this is when waiting until high contrast level in being in brightness, causes the colourity that is caused by brightness to cover.Therefore, in bigger font, covered by the high luminance contrast of font along the colored artefact of font edge.Yet along with font size reduces, the brightness of font reduces, and identical colored artefact becomes (for example, for very little font, the black/white section of font disappears, and only stays the color spot of localization) that can see very significantly.
Summary of the invention discloses
The present invention is the characteristic according to the spatial frequency response of above-mentioned people's vision system, in other words, is to have this true being created of higher cutoff frequency according to brightness CSF compared with chrominance C SF.By the present invention, low-resolution image is by the separation of images of high-resolution is become brightness data and chroma data, so that carry out suitable sampling and filtering for each brightness data and chroma data, and combination brightness data and chroma data are formed after sampling.As for chroma data, carry out traditional sub sampling, obscure to avoid colour, and carry out sub-pixel sampling for brightness data simultaneously, to improve the resolution of luminance component.And, carry out high-pass filtering for the brightness data of sub-pixel sampling, so that remove the low frequency artefact that during the sub-pixel sampling of brightness data, occurs.
Provide conceptual illustration of the present invention below.On Fig. 1 and 2, each R, the G of R, the G in a display pixel 16 and B sub-pixel 10,12 and 14 and the reflection of B numerical value are by each R, G and the B numerical value of the colour element 8 (that is, 11) of image 4 sub sampling shown in Figure 1, that have high resolution.Yet sub-pixel 10,12 is different with B numerical value with each R, the G of colour element 8 (11) with 14 each R, G with B numerical value.So, according to luminance component, carry out sub-pixel sampling shown in Figure 2, so that the luminance component of each R, the G of sub-pixel 10,12 and 14 and B numerical value reflection colour element 11,13 and 15.
Embodiments of the invention comprise hypothesis, and the method and system that can work to input color image of the linearity that not too relies on filtering and it.These embodiment can directly remove these artefacts after the colored artefact of low frequency is caused by sub-pixel sampling.This is added to brightness by generation and colour is obscured on the pattern, the LPF pattern of the color content of image reaches.This be different from by utilization attach, (that is, colour control RGB) removes the colored artefact that is caused by sub-pixel sampling and finishes basic color domain.In fact, only need to offset the colored artefact of lower frequency, because the colored artefact of high frequency is because the narrower bandwidth of colored CSF and can not being seen, shown in Fig. 6 (a).
Method and system of the present invention can, not be used in the time of can not seeing colored aliasing in the luminance signal that obtains high-resolution when viewed more far away compared with the designing technique condition at display.These technology do not need to suppose that source images is that text or image are monochromatic.
Embodiments of the invention become the image of low resolution to the image transformation of high-resolution, and the error that is caused by the sub sampling processing procedure simultaneously is reduced.When the image of high-resolution was not the form of permission separating luminance and chroma data, image was transformed into such form.The color domain of many opposition is acceptable.The color domain image of opposition is cut apart, and thus luminance channel and chrominance channe is separated, and allows thus separately to handle.
Luminance channel is transformed into the color domain (ACD) that attaches then, and such as RGB, and the ACD luminance picture is by sub-pixel sampling, so that keep brightness data when reducing resolution.Behind sub-pixel sampling, (SPS) image of sub-pixel sampling is transformed back to the color domain (OCD) of opposition, and is divided into brightness and chrominance channe separately once more.The SPS chrominance channe that produces by this separation is then by high-pass filtering, to remove the low frequency artefact that produces during sub-pixel sampling.The SPS luminance channel typically is not corrected, so that keep original brightness data.
Chrominance channe from original image is low pass filtering, then by sub sampling, so that provide chroma data for the image of low resolution.Chrominance channe of low-pass filtering is combined with the chrominance channe of sub-pixel sampling high-pass filtering, that produce from original luminance channel then for these.The chrominance channe of these combinations is also combined with the SPS luminance channel, reduces image wrong, low resolution with formation, normally in the color domain of opposition.This mistake image that reduce, low resolution can be transformed the color domain that attaches or transform to the application of wanting compatible mutually, certain other color domain.
The accompanying drawing summary
In order to obtain above-mentioned advantage and purpose with other of the present invention, will present more specific description of the present invention by with reference on accompanying drawing, show, specific embodiment of the present invention for above general introduction.Seeing that these accompanying drawings only describe typical embodiment of the present invention, rather than be looked at as limit the scope of the invention after, will describe and illustrate the present invention with additional specificity and details by using accompanying drawing, wherein:
Fig. 1 is the figure that shows for traditional image sampling of the display with three colour cell dot structures;
Fig. 2 is the figure that shows for the sub-pixel image sampling of the display with three colour cell dot structures;
Fig. 3 is the figure that shows the Utopian CSF that is transformed the numerical frequency plane;
Fig. 4 is the figure that indicator gauge is shown with the analysis of the pixel Nyquist frequency in sharp zone and sub-pixel Nyquist frequency field;
Fig. 5 shows typical preconditioning technique;
Fig. 6 (a) shows to use with pixel sampling and sub-pixel sampling frequency figure that repeat, the 1/f-power spectra analysis;
Fig. 6 (b) shows to use, figure 1/f-power spectrum and that have the improved analysis that owing to pre-service obtain that repeat with pixel sampling and sub-pixel sampling frequency;
Fig. 7 is the block scheme that shows the known use of vision mode;
Fig. 8 is the process flow diagram that shows embodiments of the invention;
Fig. 9 is the process flow diagram that shows certain embodiments of the present invention; And
Figure 10 is the figure that shows by the signal that embodiments of the invention kept.
Realize optimal mode of the present invention
By with reference to the accompanying drawings, will understand current preferred embodiment of the present invention best, parts identical on the figure are used identical numeral in the text.Figure listed above will directly be quoted as the part of this detailed description.
Will see easily that as usually describing and showing, parts of the present invention can be arranged and be designed with various different structure here on figure.Therefore, the embodiment of method and system of the present invention, following more detailed description does not plan to be used for limiting the scope of the invention, and it only is the representative of currently preferred embodiment of the present invention.
The element of embodiments of the invention can with hardware, firmware and or software implement.And, might provide the software (program) of the characteristic that is used for implementing embodiments of the invention by this way, so that software is stored in the computer-readable medium.The example of such medium, except such as information-storing device (semiconductor memory, floppy disk, hard disk etc.) and beyond the such recording medium of light storage device (CD-ROM, DVD etc.), be included in computer network (LAN, WAN, such as internet etc., and wireless communication networks) communication medium (optical fiber, wireless communication line or the like) that uses in the system.And, might realize the characteristic of embodiments of the invention with the Computer signal of concrete manifestation in electric transmission.Though the exemplary embodiment of Jie Shiing may only be described a kind of form in these forms, should see that those skilled in the art can implement these elements with any form of these forms, and still belong to scope of the present invention here.
Embodiments of the invention can be described and requirement obtains protection of the patent right according to term " monochrome ".As being used together with image in this explanation and appended claims, this term is meant that does not have the image that the colour that can see changes.Monochrome image can be the image that only comprises an one deck or a chrominance channel, or has multilayer or a plurality of chrominance channel but each color layer is identical, produces the image of solid color thus.
Embodiments of the invention can be according to " RGB " image or territory, or " color domain that attaches " or " coloured image that attaches " describes and requirement obtains protection of the patent right.As what be used in this explanation and relevant claim, these terms can refer to have any form in many component images territory of whole brightness and chrominance information, include but not limited to various RGB territory and CMYK territory.
Embodiments of the invention can be according to " YCrCb " image or territory, " colour of opposition " territory, image or channel, or " color difference " territory, image or channel is described and requirement obtains protection of the patent right.These terms as what be used, can refer to have any form in the many component images territory that comprises luminance channel that at least one is different and chrominance channe in this explanation and relevant claim, include but not limited to YCrCb, LAB, YUV and YIQ territory.
Embodiments of the invention can be used for the image transformation of high-resolution is become to have image few visible mistake, low resolution in converted image.Have more high-resolution image though these embodiment typically are used for conversion together with display device compared with display, resolution is dropped to the spendable resolution of display, other application items also are applicable.
The image that is transformed by embodiments of the invention can have various forms.When the processing procedure of these forms and embodiments of the invention when being incompatible, image can be transformed into compatible form before processing, and ought can be transformed back to original form more in case of necessity after processing.
Embodiments of the invention can illustrate that Fig. 8 shows the figure that summarizes exemplary embodiment with reference to Fig. 8.The image of this processing procedure 70 from the color domain (OCD) that is present in opposition begins, such as YCrCb, and LAB, YUV, YIQ or similar territory.When image was present in the color domain (ACD) that attaches, such as RGB or CMYK territory or certain other color space, image can be transformed into the color domain of opposition before utilizing embodiments of the invention to handle.Before some embodiment is included in and handles image transformation is become the step of compatible form.
In case image is in the color domain 72 of opposition, has different luminance channels and chrominance channe, image just " is cut apart " (step 74), so that get ready for separate processes brightness and chrominance channe." cut apart " (step 74) and can comprise sampling or filtering original OCD image 72, or brightness and chroma data and original image 72 separated additive methods.Cut apart also and can comprise image transformation.
After cutting apart, initial luminance channel 76 is transformed into (step 78) ACD luminance picture, such as the RGB image.After this step is finished, make it possible to the form that finally is shown or the territory luminance picture of sampling.In case luminance picture is transformed (step 78), just image is carried out sub-pixel sampling (step 80), so that improve the resolution of the image of the low resolution that finally obtains.Like this, the brightness data of each pixel in succession in the image of original high-resolution is assigned to each the corresponding sub-pixel in the image at low resolution.
When sub-pixel sampling (step 80) is finished, produce (SPS) luminance picture of sub-pixel sampling, this SPS luminance picture is transformed into (step 82) OCD image, and it can be called as the SPS-OCD luminance picture.This conversion is performed, so that the SPS luminance picture further is divided into (step 84) different brightness and chrominance channe.SPS luminance channel 86 typically is left and is not disturbed, until later on and till other channels combined (step 88).Yet, SPS chrominance channe 90 and 92 filtered before further combination.
These SPS chrominance channes 90 and 92 can be divided into red-green channel 90 and blue-yellow channel 92.These channels typically comprise the Cr and the Cb channel of YCrCb image, the LAB image " a " and " b " channel, the U of YUV image and V channel, the I of YIQ image and Q channel, or the similar channel in other color spaces or territory.These chrominance channes 90 and 92 are removed the low frequency artefact that occurs by high-pass filtering (step 94 and 96) during sub-pixel sampling.
In certain embodiments of the present invention, can carry out high-pass filtering (step 94 and 96) by fuzzy mask method.Fuzzy mask can use low pass nuclear.Typically, original image is processed with low pass nuclear, produces the pattern of the low pass of image.The pattern of this low pass is deducted from original unfiltered image subsequently, and keeps the intermediate value of image simultaneously.Successful embodiment has used has about 0.3 pixel to nuclear about 0.8 pixel and Gauss's low pass value.0.6 pixel be considered to successful especially with value, and cause ending on the frequency domain of about 0.168 cycle/pixel.This provides good fuzzy mask filter.Provide derivation below for gaussian kernel.
The one dimension Gaussian function of Shi Yonging is given in certain embodiments:
F ( x ) = 1 σ 2 π e - x 2 / 2 σ 2 μ = 0 - - ( 1 )
The Fourier transform of this function is given:
F ( k ) = e - 2 π 2 k 2 σ 2 - - ( 2 )
Here, we see that the σ in spatial domain (is unit with the pixel) is corresponding to 1/ π in the frequency domain 2σ (is unit with cycle/pixel).This relational expression can be used for helping determining the cutoff frequency of the wave filter of given its σ, otherwise or, determine the space σ of the fuzzy mask of given frequency, this can be directed by the CSF model.
The two-dimensional Gaussian function of Shi Yonging is given in certain embodiments:
F ( x , y ) = 1 2 π σ x σ y e - ( x 2 2 σ x 2 + y 2 2 σ y 2 ) μ x , μ y = 0 - - ( 3 )
Because can carrying out rectangular coordinate, separates Gaussian function, so the frequency response of two-dimensional Gaussian function is similar to formula (2), and when considering the importance of σ, just, the σ in the time domain xBe 1/ π in the frequency domain 2σ x, and the σ in the time domain yBe 1/ π in the frequency domain 2σ y
Successful embodiment of the present invention adopts the Gaussian Blur mask filter of executing with examining of size 3 * 3, and the numerical value of itself and value (sigma) is selected as 0.6, causes being about the cutoff frequency of the low-pass filter of 0.2 cycle/pixel.
Other embodiment of the present invention can use the Hi-pass filter of the contrary CSF of the chrominance channel that is equivalent to each opposition.These CSF can be transformed into the numeric field of cy/pix from the territory (they are modelings therein) of cy/deg.Actual conversion process is considered viewing distance, and allow for have specific display resolution (in pixel/millimeter) and different viewing distance expection or plan different should be used for customized.Because method of the present invention, when when watching more nearby unlike the viewing distance of design, colored artefact will be seen.Yet brightness resolution will be modified.
For chrominance channe 90 and 92 or for the channel of selecting, can be according to artifactitious quantity or the intensity in specific sampling processing process, introduced, or carry out this filtering (step 94 and 96) according to some other criterion.
Original OCD chrominance channe 98 and 100 low-pass filtering (step 102 and 104) can be carried out simultaneously with the processing in luminance channel 105, or can carry out at some other times.Carry out low-pass filtering (step 102 and 104) for the OCD chrominance channe, so that remove main colour frequency greater than display picture element Nyquist frequency.Therefore, these channels can be in a conventional manner with 1: 3 the factor by sub sampling (step 101 and 103), and in color channel 110, do not produce colored aliasing.
In case filtering operation is finished, the channel of separation just can be combined.The combination of chrominance channel will depend on employed color domain and change.In this exemplary embodiment, high-pass filtering, sub-pixel sampling blue-yellow (HPFSPS-B/Y) chrominance channel 97 will with blue-yellow (LPFSS-B/Y) chrominance channel 109 combined (steps 106) of low-pass filtering, traditional sub sampling, with form single height filtering, (HLF) B/Y chrominance channel 111.High-pass filtering, sub-pixel sampling red-green (HPFSPS-R/G) channel 95 also with red-green (LPFSS-R/G) channel 107 combined (steps 108) of low-pass filtering, traditional sub sampling, with form single height filtering, (HLF) R/G channel 114.
Should be pointed out that embodiments of the invention method can other variations of chrominance channel that can comprise other chrominance channels and other quantity and luminance channel, be used in other color spaces and the territory.The HLF chrominance channe 111 of combination and 114 can also with SPS luminance channel 86 combined (step 88), to form the OCD image 116 of low resolution.The OCD image 116 of low resolution when needs are used for various uses, can be transformed into other picture formats or territory then.
The method and system of these embodiment provides the image with the artifactitious low resolution of few visible colour.
In addition, above embodiment can be corrected in every way.For example, in some cases, chrominance channe 98 and 100 low-pass filtering (step 102 and 104) can be omitted.And, might form the image 116 of low resolution by only using divided luminance channel 76.In other words, might carry out each step of luminance channel 105 with respect to luminance channel 76, and omit each step of color channel 110 and with respect to the step 106 and 108 of chrominance channe 98 and 100, so that combination S PS luminance channel 86 and HPFSPS-R/G channel 95 and HPFSPS-B/Y channel 97.Like this, might form the image 116 of low resolution.
By reference Fig. 9, specific exemplary embodiment of the present invention can be described.This certain embodiments can be used to handle the RGB image of high-resolution, is used for showing on the display device of low resolution.The RGB image 120 of high-resolution can be according to the concrete needs of user or application, by randomly pre-service (step 122).Pre-service (step 122) can comprise hint (hinting), the type of low-pass filtering or some other treatment technology.Pre-service (step 122) also can be together by bypass.
After any pre-service (step 122), RGB can be transformed the color domain image of (step 124) opposition, such as LAB, YCrCb, YIQ, YUV or other image areas.In this example, use the LAB image area.In case be transformed this territory, image just can be cut apart independent L, a and the b channel that (step 126) becomes this territory, so that handle these channels dividually.Like this, colourity and luminance channel can be handled with being separated.
" L " channel 127 is transformed back to RGB territory (step 128) then, so that it can be sampled with its last display format.This conversion can comprise L layer or channel are copied into three identical R, G and B layer simply.Also can use single layer, yet actual transform method will depend on selected color transformation.
Then, the RGB luminance picture is carried out sub-pixel sampling (step 130) hereto, to keep the horizontal luminance resolution of original RGB image 120.Behind sub-pixel sampling, the image of having sampled is transformed into the color domain of (step 132) opposition once more, such as LAB.This LAB image of having sampled is cut apart (step 134), so that separate brightness and chrominance channe, for further handling chrominance channe.Here, typically do not handle luminance channel 136, so that keep original brightness data.Yet the chrominance channe 150 and 152 of sub-pixel sampling and image that cut apart is by high-pass filtering (step 146 and 148), to remove the colored aliasing of the low frequency that occurs during sub-pixel sampling.
As in other embodiment of above explanation, this high-pass filtering can be performed with the fuzzy mask filter that uses Gauss's low pass nuclear.In the embodiment of this method of use, chrominance channe is filtered, produce the chromatic diagram picture of low-pass filtering, and from SPS-RGB chromatic diagram picture, deduct the chromatic diagram picture of this low-pass filtering, to produce (HPF) SPS chromatic diagram picture or channel 147 and 149 of " high pass " filtering.High-pass filtering (step 146 and 148) typically for " a " and " b " channel carries out, but can be for a channel execution when condition permit.
For original " a " and " b " and chrominance channe 154 and 156 low-pass filtering (step 138 and 140) can with " L " processing of channel carries out simultaneously, or can carry out at some other times.For " a " and " b " chrominance channe execution low-pass filtering (step 138 and 140), to remove main colour frequency greater than display picture element Nyquist frequency.After low-pass filtering (step 138 and 140), these channels can be in a conventional manner with 1: 3 the factor by sub sampling (step 142 and 144), and do not produce colored aliasing.
Filtered and when being sampled when channel, they are combined, and have the lower resolution image of few mistake with formation.The brightness of high-pass filtering " a " channel 147 will with sub sampling, low-pass filtering " a " channel 143 combined (step 160), handled to form " a " channel 164.The brightness of high-pass filtering " b " channel 149 will with sub sampling, low-pass filtering " b " channel 145 combined (step 158), handled to form " b " channel 162.These chrominance channes 162 and 164 then with SPS luminance channel 136 combined (step 166), the LAB image 168 of, low resolution that mistake reduces to form.
The image that this mistake reduces can be transformed (step 170) RGB territory, the RGB image 172 of, low resolution that mistake reduces to produce, and it can be output to display or other devices.
The function of the processing of embodiments of the invention can be illustrated with reference to Figure 10, and Figure 10 shows with respect to brightness CSF 180 and colored CSF 182 maintained signals.The colour signal that is retained comprises that for colored CSF be undetectable high pass zone 184, and the low pass zone 186 that comprises the useful color content of image.Ideally, because the limited bandwidth of RG and BY CSF can not be seen for the observer from the frequency that this low pass colour 186 is lost.HPF colour signal 184 is the colored aliasings that carry effective monochrome information.Should be pointed out that present technique will act on coloured image because the low frequency chromatic information is held.Figure 10 is presented at not to be had overlappingly between these two colour signals, but depend on the wave filter of employed reality, and overlapping is possible.Other embodiment can comprise the use of permission overlapping wave filters as shown in figure 10, high pass 184 and low pass 186 colour signals.Overlapping can be that cost allows bigger colored bandwidth with colored aliasing.
The present invention can be embodied with other specific forms, and does not deviate from its spirit or intrinsic propesties.Described embodiment only is looked at as illustrative rather than restrictive aspect all.So scope of the present invention is represented by appended claims, rather than is represented by above-mentioned explanation.The meaning and the interior all changes of scope that belong to the equivalence of claim will be included in their scope.

Claims (20)

1. one kind is used for the image transformation of high-resolution being become to have the method for image of the low resolution of the mistake of seeing that reduces, and described method comprises following action:
The color domain of the opposition of high-resolution (OCD) image segmentation is become original intensity channel and one or more initial chroma channel separately;
Carry out sub-pixel sampling for described original intensity channel, thereby produce a color domain that attaches (ACD) luminance picture;
Described ACD luminance picture is transformed into the OCD luminance picture;
Described OCD luminance picture is divided into luminance channel and one or more SPS chrominance channe of own sub-pixel sampling (SPS) separately;
The described one or more SPS chrominance channes of high-pass filtering; And
Combined the SPS chrominance channe of described one or more high-pass filterings and described SPS luminance channel, to form the image of the low resolution that mistake reduces.
2. equipment of image that is used for the image transformation of high-resolution is become to have the low resolution of the mistake of seeing that reduces, described equipment comprises:
Be used for the color domain of the opposition of high-resolution (OCD) image segmentation is become the original intensity channel separately and the device of one or more initial chroma channels;
Thereby be used for carrying out the device that sub-pixel sampling produces color domain (ACD) luminance picture that attaches for described original intensity channel;
Be used for described ACD luminance picture is transformed into the device of OCD luminance picture;
Be used for described OCD luminance picture is divided into the luminance channel of own sub-pixel sampling (SPS) separately and the device of one or more SPS chrominance channes;
The Hi-pass filter that is used for the described one or more SPS chrominance channes of high-pass filtering; And
Be used for the combined device of the SPS chrominance channe of described one or more high-pass filterings and described SPS luminance channel with the image that forms the low resolution that mistake reduces.
One kind wherein storage be used for the image transformation of high-resolution is become to have the computer-readable medium of method of image of the low resolution of the mistake of seeing that reduces, described method comprises following action:
The color domain of the opposition of high-resolution (OCD) image segmentation is become original intensity channel and one or more initial chroma channel separately;
Carry out sub-pixel sampling for described original intensity channel, produce a color domain that attaches (ACD) luminance picture thus;
Described ACD luminance picture is transformed into the OCD luminance picture;
Described OCD luminance picture is divided into luminance channel and one or more SPS chrominance channe of sub-pixel sampling (SPS) separately;
The described one or more SPS chrominance channes of high-pass filtering; And
Combined the SPS chrominance channe of described one or more high-pass filterings and described SPS luminance channel, to form the image of the low resolution that mistake reduces.
4. one kind is used for the image transformation of high-resolution being become to have the method for image of the low resolution of the mistake of seeing that reduces, and described method comprises following action:
The color domain of the opposition of high-resolution (OCD) image segmentation is become original intensity channel and initial chroma channel separately;
Carry out sub-pixel sampling for described original intensity channel, thereby produce color domain (ACD) luminance picture that attaches;
Described ACD luminance picture is transformed into the OCD luminance picture;
Described OCD luminance picture is divided into the luminance channel and the SPS chrominance channe of sub-pixel sampling (SPS) separately;
The described SPS chrominance channe of high-pass filtering;
The described initial chroma channel of low-pass filtering;
The initial chroma channel of the described filtering of sub sampling;
Combined the SPS chrominance channe of the chrominance channe of the low-pass filtering of described sub sampling and described high-pass filtering; And
Combined described chrominance channe that has made up and described SPS luminance channel, to form the image of the low resolution that mistake reduces.
5. the method for claim 4, the image of wherein said high-resolution is a color domain image that attaches, it was transformed into the color domain image of an opposition before the described OCD image of cutting apart a high-resolution.
6. the method for claim 4, the wherein said color domain image that attaches is the RGB image.
7. the method for claim 4, the color domain image of wherein said opposition is the YCrCb image.
8. the method for claim 4, the color domain image of wherein said opposition is the LAB image.
9. the method for claim 4, wherein fast high-pass filtering comprises fuzzy mask filtering.
10. the method for claim 4, wherein said high-pass filtering comprises following action:
Come the described SPS chrominance channe of filtering by fuzzy mask filter, cause forming low pass SPS chrominance channe with Gauss's low pass nuclear; And
From described SPS chrominance channe, deduct described SPS low pass chrominance channe, to produce the SPS chrominance channe of high-pass filtering.
11. the method for claim 4, wherein said chrominance channe comprise red-green channel and blue-yellow channel.
12. the method for claim 4, wherein said chrominance channe comprise the Cr and the Cb channel of YCrCb image.
13. the method for claim 4, wherein said chrominance channe comprise the CIELab image " a " and " b " channel.
14. the method for claim 4, the image transformation that also comprises the low resolution that described mistake is reduced becomes the action of RGB image.
15. the method for claim 4, the action of wherein said execution sub-pixel sampling comprise described original intensity channel conversion is become a color domain that attaches (ACD) luminance picture, and the described ACD luminance picture of sampling.
16. a method that is used for showing with the low resolution with the mistake of seeing that reduces the image of high-resolution, described method comprises following action:
The RGB image transformation of high-resolution is become color domain (OCD) image of the opposition of high-resolution;
The OCD image segmentation of described high-resolution is become original intensity channel and initial chroma channel separately;
Described original intensity channel conversion is become the RGB luminance picture;
Carry out sub-pixel sampling for described RGB luminance picture;
(SPS) RGB luminance picture of described sub-pixel sampling is transformed into the SPS-OCD luminance picture;
Described SPS-OCD luminance picture is divided into SPS brightness and SPS chrominance channe separately;
The described SPS chrominance channe of high-pass filtering;
The described initial chroma channel of the OCD image of the described high-resolution of low-pass filtering;
The initial chroma channel of the described filtering of sub sampling;
Combined the SPS chrominance channe of the chrominance channe of the low-pass filtering of described sub sampling and described high-pass filtering;
Combined described chrominance channe that has made up and described SPS luminance channel, to form the image of the OCD low resolution that mistake reduces; And
The image transformation of the OCD low resolution that described mistake is reduced becomes the RGB image of the wrong low resolution that reduces.
17. one kind is used for the image transformation of high-resolution being become to have the method for image of the low resolution of the mistake of seeing that reduces, said method comprising the steps of:
The color domain of the opposition of high-resolution (OCD) image segmentation is become original intensity channel and initial chroma channel separately;
Carry out sub-pixel sampling for described original intensity channel, thereby produce a color domain that attaches (ACD) luminance picture;
Described ACD luminance picture is transformed into the OCD luminance picture;
Described OCD luminance picture is divided into the luminance channel and the SPS chrominance channe of sub-pixel sampling (SPS) separately;
The described SPS chrominance channe of high-pass filtering;
The described initial chroma channel of low-pass filtering;
The initial chroma channel of the described filtering of sub sampling;
Combined the SPS chrominance channe of the chrominance channe of the low-pass filtering of described sub sampling and described high-pass filtering; And
Combined described chrominance channe that has made up and described SPS luminance channel, to form the image of the low resolution that mistake reduces.
18. one kind is used for the image transformation of high-resolution being become to have the system of image of the low resolution of the mistake of seeing that reduces, described system comprises:
First dispenser is used for the color domain of the opposition of high-resolution (OCD) image segmentation is become original intensity channel and initial chroma channel separately;
The sub-pixel sampling device is used for carrying out sub-pixel sampling for described original intensity channel, thereby produces a color domain that attaches (ACD) luminance picture;
Transducer is used for described ACD luminance picture is transformed into the OCD luminance picture;
Second dispenser is used for described OCD luminance picture is divided into the luminance channel and the SPS chrominance channe of sub-pixel sampling (SPS) separately;
Hi-pass filter is used for the described SPS chrominance channe of high-pass filtering;
Low-pass filter is used for the described initial chroma channel of low-pass filtering;
Subsampler is used for the initial chroma channel of the described filtering of sub sampling;
First combiner is used to make up the chrominance channe of low-pass filtering of described sub sampling and the SPS chrominance channe of described high-pass filtering; And
Second combiner, be used for described chrominance channe that has made up and described SPS luminance channel combined, to form the image of the low resolution that mistake reduces.
19. one kind comprises the computer-readable medium of instruction of image that are used for the image transformation of high-resolution is become to have the low resolution of the mistake that reduces, described instruction comprises following action:
The color domain of the opposition of high-resolution (OCD) image segmentation is become original intensity channel and initial chroma channel separately;
Carry out sub-pixel sampling for described original intensity channel, thereby produce a color domain that attaches (ACD) luminance picture;
Described ACD luminance picture is transformed into the OCD luminance picture;
Described OCD luminance picture is divided into the luminance channel and the SPS chrominance channe of sub-pixel sampling (SPS) separately;
The described SPS chrominance channe of high-pass filtering;
The described initial chroma channel of low-pass filtering;
The initial chroma channel of the described filtering of sub sampling;
Combined the SPS chrominance channe of the chrominance channe of the low-pass filtering of described sub sampling and described high-pass filtering; And
Combined described chrominance channe that has made up and described SPS luminance channel, to form the image of the low resolution that mistake reduces.
20. a computer data signal that embodies in electric transmission, described signal have the function of image that the image transformation of high-resolution is become to have the low resolution of the mistake of seeing that reduces, described signal comprises following instruction:
The color domain of the opposition of high-resolution (OCD) image segmentation is become original intensity channel and initial chroma channel separately;
Carry out sub-pixel sampling for described original intensity channel, thereby produce a color domain that attaches (ACD) luminance picture;
Described ACD luminance picture is transformed into the OCD luminance picture;
Described OCD luminance picture is divided into the luminance channel and the SPS chrominance channe of sub-pixel sampling (SPS) separately;
The described SPS chrominance channe of high-pass filtering;
The described initial chroma channel of low-pass filtering;
The initial chroma channel of the described filtering of sub sampling;
Combined the SPS chrominance channe of the chrominance channe of the low-pass filtering of described sub sampling and described high-pass filtering; And
Combined described chrominance channe that has made up and described SPS luminance channel, to form the image of the low resolution that mistake reduces.
CN 01822601 2000-12-12 2001-12-12 Method and system for improving display resolution in images using sub-pixel sampling and visual error filtering Expired - Fee Related CN1267882C (en)

Applications Claiming Priority (6)

Application Number Priority Date Filing Date Title
US09/735,425 US6807319B2 (en) 2000-06-12 2000-12-12 Methods and systems for improving display resolution in achromatic images using sub-pixel sampling and visual error filtering
US09/735,454 US6775420B2 (en) 2000-06-12 2000-12-12 Methods and systems for improving display resolution using sub-pixel sampling and visual error compensation
US09/735,454 2000-12-12
US09/735,424 US6608632B2 (en) 2000-06-12 2000-12-12 Methods and systems for improving display resolution in images using sub-pixel sampling and visual error filtering
US09/735,424 2000-12-12
US09/735,425 2000-12-12

Publications (2)

Publication Number Publication Date
CN1489760A true CN1489760A (en) 2004-04-14
CN1267882C CN1267882C (en) 2006-08-02

Family

ID=27419197

Family Applications (1)

Application Number Title Priority Date Filing Date
CN 01822601 Expired - Fee Related CN1267882C (en) 2000-12-12 2001-12-12 Method and system for improving display resolution in images using sub-pixel sampling and visual error filtering

Country Status (5)

Country Link
EP (1) EP1350221A2 (en)
JP (1) JP2004524729A (en)
CN (1) CN1267882C (en)
TW (1) TW558899B (en)
WO (1) WO2002048960A2 (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101453543B (en) * 2007-12-03 2012-07-25 村田机械株式会社 Image processing apparatus
CN103559859A (en) * 2010-03-15 2014-02-05 夏普株式会社 Display device and television receiver
CN106470335A (en) * 2015-08-17 2017-03-01 西安诺瓦电子科技有限公司 Image processing method based on sub-pix sampling and method for displaying image

Families Citing this family (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7110012B2 (en) * 2000-06-12 2006-09-19 Sharp Laboratories Of America, Inc. System for improving display resolution
US6927804B2 (en) * 2002-09-09 2005-08-09 Eastman Kodak Company Reducing color aliasing artifacts from color digital images
US7084906B2 (en) * 2002-10-15 2006-08-01 Eastman Kodak Company Reducing computation time in removing color aliasing artifacts from color digital images
JP4813787B2 (en) * 2003-10-17 2011-11-09 パナソニック株式会社 Image processing apparatus and method
TWI343027B (en) 2005-05-20 2011-06-01 Samsung Electronics Co Ltd Display systems with multiprimary color subpixel rendering with metameric filtering and method for adjusting image data for rendering onto display as well as method for adjusting intensity values between two sets of colored subpixels on display to minimi
KR101364076B1 (en) * 2005-11-09 2014-02-26 코닌클리케 필립스 엔.브이. A method and apparatus processing pixel signals for driving a display and a display using the same
WO2007143340A2 (en) 2006-06-02 2007-12-13 Clairvoyante, Inc High dynamic contrast display system having multiple segmented backlight
US7592996B2 (en) 2006-06-02 2009-09-22 Samsung Electronics Co., Ltd. Multiprimary color display with dynamic gamut mapping
US8724928B2 (en) * 2009-08-31 2014-05-13 Intellectual Ventures Fund 83 Llc Using captured high and low resolution images
WO2016154336A1 (en) 2015-03-25 2016-09-29 Dolby Laboratories Licensing Corporation Chroma subsampling and gamut reshaping

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5170152A (en) * 1990-12-14 1992-12-08 Hewlett-Packard Company Luminance balanced encoder
EP0772179A3 (en) * 1995-10-31 1998-10-21 Cirrus Logic, Inc. Method and apparatus for determining chrominance components
US6396505B1 (en) * 1998-10-07 2002-05-28 Microsoft Corporation Methods and apparatus for detecting and reducing color errors in images

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101453543B (en) * 2007-12-03 2012-07-25 村田机械株式会社 Image processing apparatus
CN103559859A (en) * 2010-03-15 2014-02-05 夏普株式会社 Display device and television receiver
CN103559859B (en) * 2010-03-15 2016-04-06 夏普株式会社 Display device and television receiver
CN106470335A (en) * 2015-08-17 2017-03-01 西安诺瓦电子科技有限公司 Image processing method based on sub-pix sampling and method for displaying image
CN106470335B (en) * 2015-08-17 2019-05-17 西安诺瓦电子科技有限公司 Image processing method and image display method based on sub-pix sampling

Also Published As

Publication number Publication date
TW558899B (en) 2003-10-21
WO2002048960A2 (en) 2002-06-20
CN1267882C (en) 2006-08-02
EP1350221A2 (en) 2003-10-08
JP2004524729A (en) 2004-08-12
WO2002048960A3 (en) 2003-07-24

Similar Documents

Publication Publication Date Title
US6608632B2 (en) Methods and systems for improving display resolution in images using sub-pixel sampling and visual error filtering
EP1484743B1 (en) System for improving display resolution
Betrisey et al. 20.4: Displaced filtering for patterned displays
CN1267882C (en) Method and system for improving display resolution in images using sub-pixel sampling and visual error filtering
US8031205B2 (en) Image data set with embedded pre-subpixel rendered image
US6775420B2 (en) Methods and systems for improving display resolution using sub-pixel sampling and visual error compensation
JP3961706B2 (en) How to display high-fidelity and high-resolution digital color images on a low-resolution dot matrix display
JP2008538615A5 (en)
CN1291849A (en) Method and device for producing white color component and controlling brightness in display original set
US6807319B2 (en) Methods and systems for improving display resolution in achromatic images using sub-pixel sampling and visual error filtering
CN1571990A (en) Method of and display processing unit for displaying a colour image and a display apparatus comprising such a display processing unit
CN101770759B (en) Method and device for downsampling based on sub-pixel
US20150235393A1 (en) Image device and data processing system
US8515197B2 (en) Image resolution enhancing device and method for display panel
CN1992909A (en) Reconstruction method of picture-element color information
EP1239451B1 (en) Method of and apparatus for remapping subpixels for a color display from quad-subpixel to striped-subpixel format
KR100597156B1 (en) Methods and systems for improving display resolution in images using sub-pixel sampling and visual error filtering
JP2006031003A (en) Method and system for improving display resolution using sub-pixel sampling and visual error compensation
Kuno et al. Aliasing reduction method for color digital still cameras with a single-chip charge-coupled device
CN100378511C (en) Image display color processing method
Betrisey et al. 20.4: Displaced Filtering for Patterned Displays
Yamanaka et al. P. 56: X‐shaped Pixel Alignment in Large‐Scale Image Display Systems

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
C17 Cessation of patent right
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20060802

Termination date: 20111212