CN102801984A - Color image sensor and method for acquiring color digital images - Google Patents

Color image sensor and method for acquiring color digital images Download PDF

Info

Publication number
CN102801984A
CN102801984A CN2012102670827A CN201210267082A CN102801984A CN 102801984 A CN102801984 A CN 102801984A CN 2012102670827 A CN2012102670827 A CN 2012102670827A CN 201210267082 A CN201210267082 A CN 201210267082A CN 102801984 A CN102801984 A CN 102801984A
Authority
CN
China
Prior art keywords
color
color filter
element array
filter element
unit
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
CN2012102670827A
Other languages
Chinese (zh)
Other versions
CN102801984B (en
Inventor
陶霖密
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.)
Individual
Original Assignee
Individual
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 Individual filed Critical Individual
Priority to CN201210267082.7A priority Critical patent/CN102801984B/en
Publication of CN102801984A publication Critical patent/CN102801984A/en
Priority to PCT/CN2013/000898 priority patent/WO2014019339A1/en
Application granted granted Critical
Publication of CN102801984B publication Critical patent/CN102801984B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T3/00Geometric image transformation in the plane of the image
    • G06T3/40Scaling the whole image or part thereof
    • G06T3/4015Demosaicing, e.g. colour filter array [CFA], Bayer pattern
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N25/00Circuitry of solid-state image sensors [SSIS]; Control thereof
    • H04N25/10Circuitry of solid-state image sensors [SSIS]; Control thereof for transforming different wavelengths into image signals
    • H04N25/11Arrangement of colour filter arrays [CFA]; Filter mosaics
    • H04N25/13Arrangement of colour filter arrays [CFA]; Filter mosaics characterised by the spectral characteristics of the filter elements
    • H04N25/134Arrangement of colour filter arrays [CFA]; Filter mosaics characterised by the spectral characteristics of the filter elements based on three different wavelength filter elements

Landscapes

  • Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Spectroscopy & Molecular Physics (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Optical Filters (AREA)
  • Color Television Image Signal Generators (AREA)

Abstract

The invention relates to a color image sensor and a method for acquiring color digital images and in particular to a color filer element array. The color image sensor comprises an image sensor and a color filter piece, the color filter piece is covered on the surface of the image sensor, and the image sensor is composed by a sensing element array arrayed by a plurality of sensing elements. The color filter piece is composed by a color filter element array arrayed by a plurality of color filter elements. One color filter element corresponds to one sensing element and outputs a pixel in a color picture. The method for acquiring color digital images acquires a pixel value of a random color filter element by calculating through specific formulas. One color filter element of the color filter element array outputs one pixel of color images. The color image sensor and the method for acquiring color digital images greatly improve use ratio of the sensor, can reconfigure red, green, blue (RBG) tristimulus values at the same position from a plurality of color filter elements, and improve accuracy of color reduction.

Description

Color image sensor and obtain the method for color digital image
Technical field
The present invention relates to a kind of transducer and obtain the method for image, relate in particular to a kind of color image sensor and obtain the method for color digital image, particularly relate to a kind of color filter element array.Color image sensor based on this color filter element array can be used for digital picture, video equipments such as various digital cameras, DV.
Background technology
The digital color image transducer is the parts of the most critical in the digital image device such as digital camera, DV, is used for converting the optical imagery of lens focus into digital color image.Color digital image equipment can be divided into two kinds according to the quantity of the imageing sensor of its inside: (1) single image transducer color digital image equipment; Comprise most DV and digital camera; Be characterized in having only an imageing sensor (CCD or cmos image sensor) in these digital color image equipment; The brightness transition that this imageing sensor will shine the visible light on the transducer is the signal of telecommunication, and digitlization in follow-up processing, produces gray scale (black and white) image.For the colouring information that obtains occurring in nature produces coloured image, the surface that is employed in imageing sensor usually need cover the method for one deck chromatic color filter, forms color image sensor jointly by chromatic color filter and imageing sensor two parts.(2) the color digital image equipment of multiple image sensor; DV like 3CCD; The DV of 3CMOS etc., its device interior has an optical splitter, and the visible light that the external world is injected is divided into the red, green, blue three beams; Respectively by three independently imageing sensor (CCD or cmos image sensor) receive, and be quantified as color digital image; In addition, in the color digital image equipment, also have a kind of employing X3 (Japanese Sigma house journal) color image sensor, this transducer adopts three layer of three look structure, and promptly the imageing sensor of three plies is respectively blue, green, red transparent image transducer.The light Smalt part of extraneous incident is absorbed and digitlization by ground floor blue image transducer earlier; Second layer green image transducer absorbs green portion and the digitlization in the incident light; The 3rd layer red image transducer absorbs remaining light and digitlization, thereby produces color digital image.
As stated, the effect of digital image sensor is with the brightness digitlization that shines the visible light on the transducer, produces gray scale (black and white) image.Because a width of cloth digital picture is to arrange the pel array that forms by a plurality of pixels; Correspondingly; Imageing sensor is the sensing element array that is become by a plurality of sensing identical permutation, and wherein each sensing unit carries out digitlization to the light intensity (brightness) that receives independently, produces a pixel in the digital picture.
Color image sensor is the surface coverage chromatic color filter at imageing sensor, makes imageing sensor to carry out digitlization to the light intensity (brightness) of different colours in the visible light, produces coloured image.Imageing sensor in the color image sensor, the same with above-mentioned gray level image transducer, be the sensing element array that forms by a plurality of sensing identical permutation; Chromatic color filter in the color image sensor is to form color filter unit group pattern by color filter tuple repeated arrangement.Each color filter tuple is corresponding with the pixel in the coloured image.Therefore, different with the gray level image transducer, the quantity of the quantity of sensing unit pixel in the coloured image in the color image sensor.
Three kinds or four kinds of colors are generally arranged, like red colour filter unit, green colour filter unit, blue colour filter unit etc. in the color filter tuple.A color filter tuple is corresponding to a pixel in the color digital image, and its output valve is the tristimulus values of a pixel in the digital picture.Therefore, the chromatic color filter in the color image sensor is made up of the colour filter tuple of repeated arrangement, the redgreenblue value of a pixel of one of them color filter tuple output, and both are corresponding one by one.
Shown in the color filter tuple following table of prior art:
Figure BSA00000756862600021
Figure BSA00000756862600031
In sum, the color filter tuple of prior art has following two characteristics:
1) has 2x2 in a color filter tuple, 2x4, three looks such as 4x4 or four look colour filters unit;
2) the colour filter unit in the color filter tuple has definite arrangement mode, and this arrangement mode receives patent protection.
Color digital image transducer with a mega pixel is an example below, and the relation between color digital image transducer and color filter tuple, colour filter unit and the three of transducer unit is described.
Suppose to have the digital camera of one one mega pixel, promptly the color digital image transducer in the camera is exported the color digital image of a mega pixel (1,000x1,000 pixel); And suppose that this imageing sensor adopts three look Bayer colour filter tuples, as shown in Figure 1, then have on this transducer:
1,000,000 Bayer colour filter tuples are exported 1,000,000 pixels, wherein pixel of each color filter tuple output;
4,000,000 colour filter element array that (2,000x 2,000) colour filter unit forms.Wherein have 1,000,000 red colour filters first, 1,000,000 blue colour filters are first, and 2,000,000 green colour filters are first;
4,000,000 sensing element array that (2,000x 2,000) sensing unit forms are covered with the colour filter element array on it, corresponding one by one between the two.Pixel of per four sensings unit's output.
In sum, mainly there is following problem in existing color image sensor: one is pixel of four sensing unit output, and the utilance of sensing unit is low, has serious wasting phenomenon; Two is that four sensing units are in different positions, and the RGB tristimulus values are not to come from same position in the promptly same pixel, have small difference, therefore, can not correctly reflect the real color of occurring in nature.
Summary of the invention
Goal of the invention of the present invention is according to amphiblestroid structure in the human eye; A kind of color image sensor is provided and obtains the method for color digital image; A kind of color filter element array (Tao Shi colour filter element array) especially is provided; A corresponding sensing unit of color filter unit exports a colour element, has greatly improved the utilance of transducer; Can reconstruct the RGB tristimulus values of same position from a plurality of color filter unit, improve the accuracy of color reduction.
For solving the problems of the technologies described above, the technical scheme that the present invention adopts is:
A kind of color image sensor comprises imageing sensor and chromatic color filter, and described chromatic color filter is covered on the said image sensor surface, and said imageing sensor is that the sensing element array constitutes by a plurality of sensing identical permutation; Said chromatic color filter is that the color filter element array constitutes by a plurality of color filter identical permutation, and said color filter element array comprises the color filter unit of three kinds of colors at least, a corresponding sensing unit of color filter unit, a pixel in the output coloured image.
Preferably, said color filter element array comprises red colour filter first, green colour filter unit and blue colour filter unit at least.
Preferably, the color filter unit random alignment in the said color filter element array forms code book at random.
Preferably, the color filter unit of the wherein a kind of color in the said color filter element array is according to regularly arranged, and the color filter unit of other color is random alignment then, forms accurate code book at random.
A kind of color image sensor obtains the method for color digital image; It is characterized in that; According to code book and following color filter element array computational methods, the redgreenblue value of a pixel in the pairing digital picture of any color filter unit in the said color filter element array of calculating acquisition:
The computational methods of color filter element array:
The first, the total n (n>=3) of color filter unit plants the colour filter unit of different colours, is respectively C 1, C 2..., C n
The second, color coefficient Λ Ri, Λ Gi, Λ BiThe color C of (0<=Λ<=1) expression colour filter unit 1, C 2..., C nIn contain the redgreenblue composition, that is:
Λ RiC i=R
Λ GiC i=G
Λ BiC i=B
Wherein i representes the kind of the unit of colour filter in the colour filter element array, i.e. i=1, and 2,3 ..., n.If n=3 then has:
Λ = 1 0 0 0 1 0 0 0 1
The 3rd, be located at that the shared ratio of versicolor colour filter unit is respectively Ω in the color filter element array 1, Ω 2, Ω 3..., Ω n, then have:
Ω 123+...+Ω n=1。
The 4th; Versicolor color filter unit arranges according to regular Ψ and forms the color filter element array in the color filter element array, and the arrangement mode of all color filters units in this colour filter element array is called the code book (Codebook) of colour filter element array; Various ways is arranged; Like all colour filter unit random alignment, accurate random alignment, or repeated arrangement.
Fifth, according to the color filter array element codebook, the color element array in any position, the color filter element
Figure BSA00000756862600042
The color is
Figure BSA00000756862600044
color element set of its neighbors
Figure BSA00000756862600045
is defined as: Let is the filter element
Figure BSA00000756862600047
near a color filter yuan, and its color is
Figure BSA00000756862600048
Figure BSA00000756862600049
and
Figure BSA000007568626000410
The distance between the
Figure BSA000007568626000411
there where
Figure BSA000007568626000413
represents
Figure BSA000007568626000414
is filter element itself.Definition is owned
Figure BSA000007568626000416
The colour filter unit of (D is a constant) The set of forming does
Figure BSA000007568626000418
Neighborhood colour filter unit collection
Figure BSA000007568626000419
Total M colour filter is first, wherein
Figure BSA000007568626000420
The expression set
Figure BSA000007568626000421
In various colour filters unit subclass, M is arranged respectively C1, M C2, M C3..., M CnIndividual color filter is first, then has: M=M C1+ M C2+ M C3+ ...+M Cn
The 6th, according to the definition distance coefficient of the distance between the colour filter unit
Figure BSA000007568626000422
and
Figure BSA000007568626000423
Figure BSA000007568626000425
in the colour filter element array as follows:
Figure BSA00000756862600051
Wherein function
Figure BSA00000756862600052
has various ways; Inverse like square distance; Exponential function, or constant.
The 7th; According to code book; The signal value that the signal value that arbitrary position in the colour filter element array
Figure BSA00000756862600053
pairing sensing unit obtains obtains for arbitrary position
Figure BSA00000756862600057
pairing sensing unit in
Figure BSA00000756862600055
its neighborhood for
Figure BSA00000756862600054
its color for
Figure BSA00000756862600058
its color for
Figure BSA00000756862600059
in the then Xiang Yingde color digital image, the RGB tristimulus values of the pixel of colour filter Yuan
Figure BSA000007568626000510
institute's correspondence position can be calculated by following formula:
Figure BSA000007568626000511
Figure BSA000007568626000512
Figure BSA000007568626000513
Algorithm f wherein R, f G, f BVarious ways is arranged, like addition.
A kind of color filter element array that is used for color image sensor, said color filter element array comprise the color filter unit of three kinds of colors at least, and a color filter unit exports a pixel in the coloured image.
Preferably, said color filter element array comprises the first and blue colour filter of first, the green colour filter of red colour filter unit etc. at least.
Preferably, the color filter unit random alignment in the said color filter element array forms code book at random.
Preferably, the color filter unit of the wherein a kind of color in the said color filter element array is according to regularly arranged, and the color filter unit of other color is random alignment then, forms accurate code book at random.
A kind of color image sensor provided by the invention and obtain the method for color digital image, especially a kind of color filter element array is formed by color filter identical permutation, does not have the colour filter tuple, i.e. pixel of a colour filter unit output; Color filter unit more than three kinds or three kinds is arranged in the colour filter element array, therefore, the problem that exists these color filter units how to arrange; The tristimulus values of a pixel of a color filter unit output, but a color filter unit can only obtain a value in the tristimulus values, and therefore, two other value need obtain through certain algorithm computation from color filter unit on every side of current this color filter unit; Its essence is help mutually between the colour filter unit, when promptly exporting the tristimulus values of the A of colour filter unit, and the B of colour filter unit around it, C, D calculating that helps; During the tristimulus values of the output B of colour filter unit, the calculating that helps of A, C, the E around it.The value of each colour filter unit has all obtained recycling repeatedly.
The first display of color filter of the present invention has following characteristic:
1, color filter element array of the present invention by color filter unit according to certain ratio with certain regularly arranged composition, do not have the colour filter tuple, neither form colour filter unit group pattern by colour filter tuple repeated arrangement;
2, described color filter unit comprises three kinds or more than the colour filter unit of three kinds of colors, and the ratio between the versicolor colour filter unit can be the shades of colour equal quantities, also can be to have different ratios between each color;
3, the color filter identical permutation of described multiple color forms the queueing discipline of color filter unit display, can be certain repetitive structure, also can be that each colour filter unit is according to certain ratio, random alignment;
4, in the described color filter element array, a pixel in each color filter unit output coloured image, color filter is first, sensing is first, pixel three is corresponding one by one.
5, Fig. 9 has shown the different of two kinds of color image sensors and digital color image acquisition methods.As can be seen from the figure, the method for the present invention's proposition comprises colour filter element array and digital image reconstruction algorithm two parts.
Description of drawings
Below in conjunction with accompanying drawing and embodiment the present invention is done further detailed explanation.
Fig. 1 is a prior art color digital image sensor construction sketch map;
101 is that Bayer colour filter element array, 102 is the sensing element array among the figure.
Fig. 2 is a color digital image transducer sketch map of the present invention;
201 is Tao Shi colour filter element array of the present invention among the figure; 202 is the transducer element array
Fig. 3 is the sketch map of Tao Shi colour filter element array of the present invention.
Fig. 4 is the sketch map of Tao Shi colour filter element array of the present invention;
Wherein Figure 40 1 is colour filter unit
Figure BSA00000756862600061
Fig. 5 is the sketch map of Tao Shi colour filter element array of the present invention.
Fig. 6 is the sketch map of Tao Shi colour filter element array of the present invention.
Fig. 7 is 401 a neighborhood colour filter unit collection example one among Fig. 4;
Numeral among the figure is used to mark different colour filter units.
Fig. 8 is 401 a neighborhood colour filter unit collection example two among Fig. 4;
Numeral among the figure is used to mark different colour filter units.
Fig. 9 a obtains schematic diagram for the prior art digital picture.
Fig. 9 b obtains schematic diagram for the digital picture that the present invention proposes.
Embodiment
Embodiment 1
As shown in Figure 2, a kind of color image sensor comprises imageing sensor and chromatic color filter, and described chromatic color filter is covered on the said image sensor surface, and said imageing sensor is that the sensing element array constitutes by a plurality of sensing identical permutation; Said chromatic color filter is that the color filter element array constitutes by a plurality of color filter identical permutation; Said color filter element array comprises the color filter unit of three kinds of colors at least; A corresponding sensing unit of color filter unit, a pixel in the output coloured image.Preferably, said color filter element array comprises red colour filter first, green colour filter unit and blue colour filter unit, like Fig. 2, Fig. 3, Fig. 4 and shown in Figure 5.Color filter unit random alignment in the said color filter element array, as shown in Figure 2.Wherein a kind of color color filter unit in the said color filter element array is according to regularly arranged, and the color filter unit of other color is random alignment then, like Fig. 4, Fig. 5 and shown in Figure 6.
Embodiment 2
A kind of method of utilizing above-mentioned color image sensor to obtain color digital image is calculated in the following manner and is obtained the first pixel value
Figure BSA00000756862600071
of any color filter
Colour vision research shows that the combination of multiple color photosensitive unit all can produce the redgreenblue coloured image.The present invention proposes the computational methods of color filter element array:
(1) the total n (n>=3) of color filter unit plants the colour filter unit of different colours, is respectively C 1, C 2..., C n
(2) color coefficient Λ Ri, Λ Gi, Λ BiThe color C of (0<=Λ<=1) expression colour filter unit 1, C 2..., C nIn contain the redgreenblue composition, that is:
Λ RiC i=R
Λ GiC i=G
Λ BiC i=B
Wherein i representes the kind of the unit of colour filter in the colour filter element array, i.e. i=1, and 2,3 ..., n.If n=3 then has:
Λ = 1 0 0 0 1 0 0 0 1
(3) be located at that the shared ratio of versicolor colour filter unit is respectively Ω in the color filter element array 1, Ω 2, Ω 3..., Ω n, then have:
Ω 123+...+Ω n=1;
(4) versicolor color filter unit arranges according to regular Ψ and forms the color filter element array in the color filter element array; The arrangement mode of all color filter units in this colour filter element array; Be called the code book (Codebook) of colour filter element array, various ways is arranged, like all colour filter unit random alignment; Accurate random alignment, and repeated arrangement etc.
(5) According to the color filter array element codebook, the color element array in any position, the color filter element
Figure BSA00000756862600073
The color is
Figure BSA00000756862600075
color element set of its neighbors
Figure BSA00000756862600076
is defined as: Let
Figure BSA00000756862600077
is the filter element
Figure BSA00000756862600078
near a color filter yuan, and its color is
Figure BSA00000756862600079
Figure BSA000007568626000710
and
Figure BSA000007568626000711
The distance between the
Figure BSA000007568626000712
there
Figure BSA000007568626000713
where
Figure BSA000007568626000714
represents
Figure BSA000007568626000715
is filter element itself.Definition is owned
Figure BSA000007568626000717
The colour filter unit of (D is a constant)
Figure BSA000007568626000718
The set of forming does
Figure BSA000007568626000719
Neighborhood colour filter unit collection
Figure BSA000007568626000720
Total M colour filter is first, wherein
Figure BSA000007568626000721
The expression set In various colour filters unit subclass, M is arranged respectively C1, M C2, M C3..., M CnIndividual color filter is first, then has: M=M C1+ M C2+ M C3+ ...+M Cn
(6) According to the color filter element array element
Figure BSA00000756862600081
and
Figure BSA00000756862600082
the distance between the
Figure BSA00000756862600083
define the distance coefficient
Figure BSA00000756862600084
is as follows:
Figure BSA00000756862600085
Wherein function
Figure BSA00000756862600086
has various ways; Inverse like square distance; Exponential function, and constant etc.
(7) according to code book; The signal value that the signal value that arbitrary position in the colour filter element array
Figure BSA00000756862600087
pairing sensing unit obtains obtains for arbitrary position
Figure BSA000007568626000811
pairing sensing unit in its neighborhood
Figure BSA000007568626000810
for
Figure BSA00000756862600088
its color for
Figure BSA000007568626000812
its color for
Figure BSA000007568626000813
in the then Xiang Yingde color digital image, the pixel value
Figure BSA000007568626000815
Figure BSA000007568626000816
of colour filter Yuan
Figure BSA000007568626000814
institute's correspondence position can be calculated by following formula:
Figure BSA000007568626000817
Figure BSA000007568626000819
Algorithm f wherein R, f G, f BVarious ways is arranged, like addition etc.
Color filter element array model one:
Fig. 2 has showed a color filter element array color digital image transducer that the present invention proposes, and color filter unit display is wherein formed according to certain ratio random alignment by redgreenblue colour filter unit.
Color filter element array model two:
Human retina research shows the variable-scale of redgreenblue photosensory cell in the retina.Fig. 3 has showed the color filter element array that a present invention proposes, and display Smalt colour filter unit of color filter unit wherein accounts for 1/9, and green and red colour filter unit respectively accounts for 4/9, and each color filter unit is according to evenly distributed regularly arranged.
Color filter element array model three:
Human retina research shows the variable-scale of redgreenblue photosensory cell in the retina.Fig. 4 has showed a color filter element array that the present invention proposes; Display Smalt colour filter unit of color filter unit wherein accounts for 1/9; Green and red colour filter unit respectively accounts for 4/9; Each color filter unit is regularly arranged according to accurate random alignment, and promptly blue colour filter unit is regularly arranged in the colour filter element array, red and green colour filter unit random alignment in the colour filter element array.
Color filter element array model four:
Computer vision research shows, the ratio of redgreenblue photosensitive unit near the time, color resolution is higher in the coloured image of generation.Fig. 5 has showed a color filter element array that the present invention proposes; Display Smalt colour filter unit of color filter unit wherein accounts for 2/9; Green and red colour filter unit respectively accounts for 3.5/9; Each color filter unit is regularly arranged according to accurate random alignment, and promptly blue colour filter unit is regularly arranged in the colour filter element array, red and green colour filter unit random alignment in the colour filter element array.
Color filter element array model five:
Computer vision research shows, the ratio of the blue or green four color sensation light unit of RGB near the time, in the coloured image of generation the reduction of color well, image resolution ratio is high.Fig. 6 has showed a color filter element array that the present invention proposes; Display Smalt colour filter unit of color filter unit wherein accounts for 1/9; Cyan colour filter unit accounts for 1/9, and green and red colour filter unit respectively accounts for 3.5/9, and each color filter unit is regularly arranged according to accurate random alignment; It is regularly arranged in the colour filter element array, red, green, the cyan colour filter unit random alignment in the colour filter element array of blue colour filter unit.
Based on above-mentioned color filter element array model, the algorithm that is produced coloured image by the color filter element array is following:
Filter element array according to a code book, the color element array in any position, the color filter element
Figure BSA00000756862600091
The color is
Figure BSA00000756862600092
color element set of its neighbors is defined as: Let
Figure BSA00000756862600095
is the filter element near a color filter element, and its The color is
Figure BSA00000756862600098
and
Figure BSA00000756862600099
The distance between the
Figure BSA000007568626000910
there
Figure BSA000007568626000911
where represents
Figure BSA000007568626000913
is
Figure BSA000007568626000914
filter element itself.Definition is owned
Figure BSA000007568626000915
The colour filter unit of (D is a constant) The set of forming does
Figure BSA000007568626000917
Neighborhood colour filter unit collection Total M colour filter is first, wherein
Figure BSA000007568626000919
The expression set
Figure BSA000007568626000920
In various colour filters unit subclass, M is arranged respectively C1, M C2, M C3..., M CnIndividual color filter is first, then has: M=M C1+ M C2+ M C3+ ...+M Cn
According to the color filter element array element
Figure BSA000007568626000921
and
Figure BSA000007568626000922
the distance between the
Figure BSA000007568626000923
define the distance coefficient
Figure BSA000007568626000924
is as follows:
Figure BSA000007568626000925
Wherein function
Figure BSA000007568626000926
has various ways; Inverse like square distance; Exponential function, and constant etc.
According to code book; The signal value that the signal value that arbitrary position in the colour filter element array pairing sensing unit obtains obtains for arbitrary position
Figure BSA000007568626000931
pairing sensing unit in
Figure BSA000007568626000929
its neighborhood
Figure BSA000007568626000930
for
Figure BSA000007568626000928
its color for
Figure BSA000007568626000932
its color for in the then Xiang Yingde color digital image, the pixel value
Figure BSA000007568626000935
of colour filter Yuan
Figure BSA000007568626000934
institute's correspondence position can be calculated by following formula:
Figure BSA000007568626000936
Figure BSA000007568626000937
Figure BSA000007568626000938
Algorithm f wherein R, f G, f BVarious ways is arranged, like addition etc.
Algorithm examples one: following is example with color filter element array model four, illustrates the aforementioned calculation method:
It is as shown in Figure 7 to get its neighborhood colour filter unit's collection
Figure BSA00000756862600102
for with the colour filter unit of 401 indications among Fig. 4, and gets:
Λ=1;
Color image can be calculated pixel corresponds to
Figure BSA00000756862600104
pixel
Figure BSA00000756862600105
color value
Figure BSA00000756862600106
is:
R X = 1 4 Σ R y G X = 1 4 Σ G y B X = B y
R in the formula y, G y, B yExpression field set respectively
Figure BSA00000756862600108
The colour filter unit collection of middle redgreenblue, promptly
Figure BSA00000756862600109
Algorithm examples two: following is example with color filter element array model four, illustrates the aforementioned calculation method:
It is as shown in Figure 8 to get its neighborhood colour filter unit's collection
Figure BSA000007568626001011
for with the colour filter unit of 401 indications among Fig. 4; According to the distance between the colour filter unit
Figure BSA000007568626001012
and
Figure BSA000007568626001013
, get respectively:
w 0 = 4 w 1 = 2 w 2 = 1
In the following formula, then
Figure BSA000007568626001016
then
Figure BSA000007568626001018
Figure BSA000007568626001019
then
Figure BSA000007568626001020
be the increase index decreased of weight with distance.Color image can be calculated pixel corresponds to
Figure BSA000007568626001021
pixel
Figure BSA000007568626001022
color value
Figure BSA000007568626001023
is:
R X = 1 12 Σ ( W xy R y ) G X = 1 10 Σ ( W xy G y ) B X = B y
R in the formula y, G y, B yExpression field set respectively
Figure BSA000007568626001025
The colour filter unit collection of middle redgreenblue, promptly
Embodiment 3
A kind of color filter element array that is used for color image sensor, said color filter element array comprise the color filter unit of three kinds of colors at least, a pixel in the color filter unit output coloured image in the said color filter element array.Preferably, said color filter element array comprises red colour filter first, green colour filter unit and blue colour filter unit, like Fig. 2, Fig. 3, Fig. 4 and shown in Figure 5.Preferably, the color filter unit random alignment in the said color filter element array, as shown in Figure 2.Preferred, one of them the color filter unit in the said color filter element array is according to regularly arranged, and other color filter unit is random alignment then, like Fig. 4, Fig. 5 and shown in Figure 6.
The foregoing description is not the exhaustive of embodiment; The embodiment that other also can be arranged; The foregoing description purpose is to explain the present invention, and unrestricted protection scope of the present invention, all application that come by simple change of the present invention all drop in protection scope of the present invention.

Claims (9)

1. a color image sensor comprises imageing sensor and chromatic color filter, and described chromatic color filter is covered on the said image sensor surface, and said imageing sensor is that the sensing element array constitutes by a plurality of sensing identical permutation; It is characterized in that: said chromatic color filter is arranged as the color filter element array by a plurality of color filter unit according to queueing discipline and constitutes; Said color filter element array comprises the color filter unit of three kinds of colors at least; A corresponding sensing unit of color filter unit, a pixel in the output coloured image.
2. color image sensor according to claim 1 is characterized in that, said color filter element array comprises red colour filter first, green colour filter unit and blue colour filter unit at least.
3. according to each described color image sensor of claim 1-2, it is characterized in that the queueing discipline of the color filter unit in the said color filter element array is a random alignment, forms code book at random.
4. according to each described color image sensor of claim 1-2; It is characterized in that; The color filter unit of the wherein a kind of color in the said color filter element array is according to regularly arranged, and the queueing discipline of the color filter of other color unit is a random alignment, forms accurate code book at random.
5. method of utilizing each described color image sensor of claim 1-4 to obtain color digital image; It is characterized in that; According to code book and following color filter element array computational methods, the redgreenblue value of a pixel in the pairing digital picture of any color filter unit in the said color filter element array of calculating acquisition:
The computational methods of color filter element array:
The first, the total n (n>=3) of color filter unit plants the colour filter unit of different colours, is respectively C 1, C 2..., C n
The second, color coefficient Λ Ri, Λ Gi, Λ BiThe color C of (0<=Λ<=1) expression colour filter unit 1, C 2..., C nIn contain the redgreenblue composition, that is:
Λ RiC i=R
Λ GiC i=G
Λ BiC i=B
Wherein i representes the kind of the unit of colour filter in the colour filter element array, i.e. i=1, and 2,3 ..., n.If n=3 then has:
Λ = 1 0 0 0 1 0 0 0 1
The 3rd, be located at that the shared ratio of versicolor colour filter unit is respectively Ω in the color filter element array 1, Ω 2, Ω 3..., Ω n, then have:
Ω 123+...+Ω n=1;
The 4th; Versicolor color filter unit arranges according to regular Ψ and forms the color filter element array in the color filter element array, and the arrangement mode of all color filters units in this colour filter element array is called the code book of colour filter element array; Various ways is arranged; Like all colour filter unit random alignment, accurate random alignment, or repeated arrangement;
Fifth, according to the color filter array element codebook, the color element array in any position, the color filter element The color is
Figure FSA00000756862500022
Figure FSA00000756862500023
color element of its neighbors Set
Figure FSA00000756862500024
is defined as: Let is the filter element
Figure FSA00000756862500026
near a color filter element, and its color is
Figure FSA00000756862500027
and
Figure FSA00000756862500029
The distance between the there
Figure FSA000007568625000211
where
Figure FSA000007568625000212
represents
Figure FSA000007568625000213
is
Figure FSA000007568625000214
filter element itself.Definition is owned
Figure FSA000007568625000215
The colour filter unit of (D is a constant)
Figure FSA000007568625000216
The set of forming does
Figure FSA000007568625000217
Neighborhood colour filter unit collection
Figure FSA000007568625000218
Total M colour filter is first, wherein
Figure FSA000007568625000219
The expression set
Figure FSA000007568625000220
In various colour filters unit subclass, M is arranged respectively C1, M C2, M C3..., M CnIndividual color filter is first, then has: M=M C1+ M C2+ M C3+ ...+M Cn
The 6th, according to the definition distance coefficient of the distance
Figure FSA000007568625000223
between the colour filter unit
Figure FSA000007568625000221
and
Figure FSA000007568625000222
Figure FSA000007568625000224
in the colour filter element array as follows:
Wherein function
Figure FSA000007568625000226
has various ways; Inverse like square distance; Exponential function, or constant;
The 7th; According to code book; The signal value that the signal value that arbitrary position in the colour filter element array
Figure FSA000007568625000227
pairing sensing unit obtains obtains for arbitrary position
Figure FSA000007568625000231
pairing sensing unit in its neighborhood
Figure FSA000007568625000230
for
Figure FSA000007568625000228
its color for
Figure FSA000007568625000232
its color Wei
Figure FSA000007568625000233
in the then middle color digital image, the RGB tristimulus values of the pixel of colour filter middle institute's correspondence position can be calculated by following formula:
Figure FSA000007568625000235
Figure FSA000007568625000237
Algorithm f wherein R, f G, f BVarious ways is arranged, like addition.
6. color filter element array that is used for color image sensor, said color filter element array comprise the color filter unit of three kinds of colors at least, it is characterized in that: a pixel in color filter unit output coloured image.
7. color filter element array according to claim 6 is characterized in that, said color filter element array comprises the first and blue colour filter of first, the green colour filter of red colour filter unit etc. at least.
8. according to claim 6 or 7 described color filter element array, it is characterized in that the color filter unit random alignment in the said color filter element array forms code book at random.
9. according to claim 6 or 7 described color filter element array, it is characterized in that the color filter unit of the wherein a kind of color in the said color filter element array is according to regularly arranged, the color filter unit of other color is random alignment then, forms accurate code book at random.
CN201210267082.7A 2012-07-31 2012-07-31 The method of color image sensor and acquisition color digital image Expired - Fee Related CN102801984B (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN201210267082.7A CN102801984B (en) 2012-07-31 2012-07-31 The method of color image sensor and acquisition color digital image
PCT/CN2013/000898 WO2014019339A1 (en) 2012-07-31 2013-07-30 Colour image sensor and method for acquiring colour digital image

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201210267082.7A CN102801984B (en) 2012-07-31 2012-07-31 The method of color image sensor and acquisition color digital image

Publications (2)

Publication Number Publication Date
CN102801984A true CN102801984A (en) 2012-11-28
CN102801984B CN102801984B (en) 2016-04-20

Family

ID=47200939

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201210267082.7A Expired - Fee Related CN102801984B (en) 2012-07-31 2012-07-31 The method of color image sensor and acquisition color digital image

Country Status (2)

Country Link
CN (1) CN102801984B (en)
WO (1) WO2014019339A1 (en)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103002292A (en) * 2012-12-26 2013-03-27 陶霖密 Colorful image sensor and colorful color-filtering cell array
WO2014019339A1 (en) * 2012-07-31 2014-02-06 Tao Linmi Colour image sensor and method for acquiring colour digital image
CN104394391A (en) * 2014-11-25 2015-03-04 广东威创视讯科技股份有限公司 Method and system for processing image data of dot matrix images acquired by camera
EP2899690A3 (en) * 2014-01-24 2015-09-09 Goodrich Corporation Random imaging
CN106231064A (en) * 2016-06-21 2016-12-14 努比亚技术有限公司 A kind of terminal and three primary colours sensor based on virtual unit drive reconstructing method
CN109900704A (en) * 2019-03-12 2019-06-18 西安交通大学 With gear wear surface microscopic topographic in-situ three-dimensional reconstructing method

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP6595161B2 (en) 2014-07-31 2019-10-23 ソニーセミコンダクタソリューションズ株式会社 Image processing apparatus, image processing method, and imaging apparatus

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1571476A (en) * 2004-04-29 2005-01-26 上海交通大学 Method of color interpolation in digital camera
CN1870282A (en) * 2005-05-25 2006-11-29 三星电机株式会社 Image sensor for semiconductor light-sensing device and image processing apparatus using the same
JP2007159141A (en) * 2005-12-07 2007-06-21 Olypmus Soft Imaging Solutions Gmbh Method for calculating color correction
US7570282B1 (en) * 1998-12-09 2009-08-04 Aptina Imaging Corporation Color correction of multiple colors using a calibrated technique
US20110013054A1 (en) * 2009-07-17 2011-01-20 Searete Llc Color filters and demosaicing techniques for digital imaging
CN102376083A (en) * 2010-07-14 2012-03-14 豪威科技有限公司 Cross-color image processing systems and methods for sharpness enhancement

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1377191A (en) * 2002-05-09 2002-10-30 北京长信嘉信息技术有限公司 Colour reconstruction method for colour filter array in digital camera
CN101939762A (en) * 2008-02-07 2011-01-05 Nxp股份有限公司 Method and device for reconstructing a color image
KR20120106762A (en) * 2009-11-20 2012-09-26 트리퓨라리 싱 Method and system for compressive color image sampling and reconstruction
CN102510450A (en) * 2011-10-17 2012-06-20 北京瑞澜联合通信技术有限公司 Image sensor, pick-up device and image data generation method
CN102801984B (en) * 2012-07-31 2016-04-20 陶霖密 The method of color image sensor and acquisition color digital image
CN103002292A (en) * 2012-12-26 2013-03-27 陶霖密 Colorful image sensor and colorful color-filtering cell array

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7570282B1 (en) * 1998-12-09 2009-08-04 Aptina Imaging Corporation Color correction of multiple colors using a calibrated technique
CN1571476A (en) * 2004-04-29 2005-01-26 上海交通大学 Method of color interpolation in digital camera
CN1870282A (en) * 2005-05-25 2006-11-29 三星电机株式会社 Image sensor for semiconductor light-sensing device and image processing apparatus using the same
JP2007159141A (en) * 2005-12-07 2007-06-21 Olypmus Soft Imaging Solutions Gmbh Method for calculating color correction
US20110013054A1 (en) * 2009-07-17 2011-01-20 Searete Llc Color filters and demosaicing techniques for digital imaging
CN102376083A (en) * 2010-07-14 2012-03-14 豪威科技有限公司 Cross-color image processing systems and methods for sharpness enhancement

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2014019339A1 (en) * 2012-07-31 2014-02-06 Tao Linmi Colour image sensor and method for acquiring colour digital image
CN103002292A (en) * 2012-12-26 2013-03-27 陶霖密 Colorful image sensor and colorful color-filtering cell array
EP2899690A3 (en) * 2014-01-24 2015-09-09 Goodrich Corporation Random imaging
US9300865B2 (en) 2014-01-24 2016-03-29 Goodrich Corporation Random imaging
CN104394391A (en) * 2014-11-25 2015-03-04 广东威创视讯科技股份有限公司 Method and system for processing image data of dot matrix images acquired by camera
CN106231064A (en) * 2016-06-21 2016-12-14 努比亚技术有限公司 A kind of terminal and three primary colours sensor based on virtual unit drive reconstructing method
CN106231064B (en) * 2016-06-21 2019-06-21 努比亚技术有限公司 A kind of terminal and the three primary colours sensor based on virtual unit drive reconstructing method
CN109900704A (en) * 2019-03-12 2019-06-18 西安交通大学 With gear wear surface microscopic topographic in-situ three-dimensional reconstructing method

Also Published As

Publication number Publication date
CN102801984B (en) 2016-04-20
WO2014019339A1 (en) 2014-02-06

Similar Documents

Publication Publication Date Title
CN102801984B (en) The method of color image sensor and acquisition color digital image
KR100982685B1 (en) System and method for a high performance color filter mosaic array
CN102204258B (en) Image inputting apparatus
CN101779288B (en) Solid-state imaging device
CN102768412B (en) Infrared imaging system and operational approach
CN106341628A (en) Image Sensor Having Wide Dynamic Range, Pixel Circuit Of The Image Sensor, And Operating Method Of The Image Sensor
JP2004228662A (en) Image pickup apparatus
KR102605761B1 (en) System and method for acquiring visible and near-infrared images by single matrix sensor
CN104717474B (en) Image processing method, module and the electronic equipment comprising it
CN109922286A (en) Cmos image sensor and its imaging method
CN111818314A (en) Filter array and image sensor
CN106911919A (en) Color image sensor and coloured image imaging method
CN105810702A (en) Optical isolation grid over color filter array
US9368530B2 (en) Image sensor and color filter array including multiple selectable meshed filter sets
CN110507283A (en) Retina camera and its implementation
US10446600B2 (en) Imaging system and imaging device having a random optical filter array
CN101150731A (en) Color filtering array for digital imaging and its imaging method
CN206559541U (en) Color image sensor
CN103546729B (en) Gray-color dual-mode TDI-CMOS image sensor and control method
CN101939762A (en) Method and device for reconstructing a color image
CN208820948U (en) Image sensing apparatus
CN103002292A (en) Colorful image sensor and colorful color-filtering cell array
CN108305883A (en) Imaging sensor
CN212785636U (en) Filter array, image sensor and application equipment thereof
CN206908722U (en) Optimization color filter array for color image sensor

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

Granted publication date: 20160420

Termination date: 20180731

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