CN104010174B - Image processing equipment and image processing method - Google Patents
Image processing equipment and image processing method Download PDFInfo
- Publication number
- CN104010174B CN104010174B CN201410201465.3A CN201410201465A CN104010174B CN 104010174 B CN104010174 B CN 104010174B CN 201410201465 A CN201410201465 A CN 201410201465A CN 104010174 B CN104010174 B CN 104010174B
- Authority
- CN
- China
- Prior art keywords
- picture signal
- target area
- signal
- frequency
- pixel
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000003672 processing method Methods 0.000 title claims abstract description 5
- 238000000034 method Methods 0.000 claims description 44
- 238000001514 detection method Methods 0.000 claims description 5
- 239000000203 mixture Substances 0.000 claims description 5
- 230000015572 biosynthetic process Effects 0.000 claims description 3
- 238000003786 synthesis reaction Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims 4
- 230000001629 suppression Effects 0.000 abstract description 13
- 230000008569 process Effects 0.000 description 32
- 230000008859 change Effects 0.000 description 9
- 238000010586 diagram Methods 0.000 description 8
- 230000004075 alteration Effects 0.000 description 6
- 230000005611 electricity Effects 0.000 description 6
- 238000006243 chemical reaction Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 239000011159 matrix material Substances 0.000 description 4
- 238000005070 sampling Methods 0.000 description 4
- 230000009467 reduction Effects 0.000 description 3
- 238000003780 insertion Methods 0.000 description 2
- 230000037431 insertion Effects 0.000 description 2
- 208000031481 Pathologic Constriction Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 description 1
- 239000010931 gold Substances 0.000 description 1
- 229910052737 gold Inorganic materials 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 230000002194 synthesizing effect Effects 0.000 description 1
- 210000001215 vagina Anatomy 0.000 description 1
Landscapes
- Color Television Image Signal Generators (AREA)
Abstract
The present invention is image processing equipment and image processing method.Purpose is the pseudo-normal complexion moire fringes occurred in suppression high-frequency region.The color interpolation circuit (300 and 350) of colour difference signal generative circuit (104) is by from being configured with multiple color filter in a predetermined pattern and being provided with the picture signal that the camera head of the pixel corresponding with each color filter obtained and separate into the picture signal of each color filter (R, G1, G2 and B), and the picture signal of each color filter is carried out interpolation processing.Then, at least one difference between false colour decision circuitry (360) picture signal based on G1 wave filter and the slope of the picture signal of G2 wave filter and G1 wave filter and G2 wave filter in the target area judges whether target area is high-frequency region.
Description
The application be filing date on October 13rd, 2009, national applications number be 200980161949.7, invention entitled
The divisional application of the PCT application (international application no is PCT/JP2009/067741) of " camera head ".
Technical field
The present invention relates to the image processing equipment of a kind of effect of noise that can reduce in high-frequency region.
Background technology
Traditionally, by using the surrounding pixel of object pixel object pixel to be carried out color interpolation, as being used for suppressing
The method of generation false colour in the high-frequency region of subject.
With reference to Figure 16, by explanation to from the redness (R) configured with bayer-pattern, green (G) and the picture of blue (B) wave filter
The picture signal of element output carries out the example of the process of color interpolation.
In figure 16, the G wave filter that will be located in the horizontal direction of R wave filter and the vertical direction of B wave filter is referred to as G1
Wave filter, and the G wave filter that will be located in the vertical direction of R wave filter and the horizontal direction of B wave filter is referred to as G2 wave filter.And
And, G1sig, G2sig, Rsig the letter exported from the pixel corresponding with G1, G2, R and B wave filter is represented respectively with Bsig
Number.
The image that picture signal is separated into the picture signal being made up of G1sig, is made up of G2sig by this image processing equipment
Signal, the picture signal being made up of Rsig and the picture signal being made up of Bsig, row interpolation of going forward side by side processes, so that in each color
Picture signal in all pixels there is signal.
Then, G1sig or G2sig of the pixel of the object pixel perpendicular abutment of image processing equipment acquisition and color interpolation
The degree of association of value and the degree of association of value of G1sig or G2sig of the pixel adjacent with object pixel level.If level side
Degree of association upwards higher than the degree of association in vertical direction, then image processing equipment using the value of Rsig-G1sig as colour difference signal
R-G, and using the value of Bsig-G2sig as colour difference signal B-G.On the contrary, if the degree of association in vertical direction is higher than horizontal direction
On degree of association, then image processing equipment is using the value of Rsig-G2sig as colour difference signal R-G, and the value of Bsig-G1sig is made
For colour difference signal B-G (seeing patent documentation 1).
Alternatively, image processing equipment is by using and the poor phase between the value of Rsig-G1sig and the value of Rsig-G2sig
Corresponding weight coefficient, using the meansigma methods of the value of Rsig-G1sig and the value of Rsig-G2sig as colour difference signal R-G.Equally,
Image processing equipment passes through to use the weight coefficient corresponding with the difference between the value of Bsig-G1sig and the value of Bsig-G2sig,
Using the meansigma methods of the value of Bsig-G1sig and the value of Bsig-G2sig as colour difference signal B-G.
By this way, by suitably selecting the pixel of color interpolation to be used for generate colour difference signal, reduce
The process of false colour.
And, when luminance signal to be generated, by according to being positioned at horizontally and vertically going up of object pixel
The degree of association of pixel and suitably select the pixel of interpolation to be used for generate picture signal, thus in suppression moire fringes (moire)
And carry out in the case of preventing the unnatural sense in whole image processing (seeing patent documentation 3).
Reference listing
Patent documentation
Patent documentation 1: Japanese Unexamined Patent Publication 2002-300590
Patent documentation 2: Japanese Unexamined Patent Publication 08-023541
Patent documentation 3: Japanese Unexamined Patent Publication 2007-336384
Summary of the invention
Technical problem
But, even if in the process above, in the picture signal included near nyquist frequency or comprise aliasing noise
Picture signal region in also fail to correctly obtain the degree of association horizontally and vertically gone up, consequently, it is possible to remain
False colour or moire fringes.False colour or the improvement leeway of moire fringes is reduced further it is taken as that exist.
For solving the scheme of problem
In order to solve the problems referred to above, include according to the image processing equipment of the present invention: interpolation component, for filling from shooting
The picture signal putting acquisition separates into the picture signal of each color filter, and by using the image of each color filter to believe
Number carry out interpolation processing, wherein, in described camera head, be configured with multiple color filter in a predetermined pattern and be provided with
The pixel corresponding with each color filter；And judgement parts, for based on the plurality of color filter in target area
The first color filter in device and the slope of the picture signal of the second color filter and described first color filter and
At least one difference between the picture signal of described second color filter, it is judged that whether described target area is high-frequency region,
Wherein, identical by the cycle and space phase configures and described first color filter and described second color in the way of offseting
The pixel that wave filter is corresponding.
The beneficial effect of the invention
According to the present invention it is possible to provide the effect of a kind of pseudo-normal complexion moire fringes that can increase and occur in suppression high-frequency region
Image processing equipment.
Accompanying drawing explanation
Fig. 1 is the block diagram of the structure illustrating image processing equipment according to a first embodiment of the present invention.
Fig. 2 is the figure illustrating the subregion being made up of in camera head 3 × 3 pixels constituted with bayer-pattern.
Fig. 3 is the block diagram of the structure illustrating colour difference signal generative circuit 104 according to a first embodiment of the present invention.
Fig. 4 is the figure for process that color interpolation circuit 300 carried out is described.
Fig. 5 includes the figure illustrating the example of digital filter.
Fig. 6 includes illustrating the figure of example of frequency characteristic of LPF301～304 and LPF351～353.
Fig. 7 is the figure of the subject illustrating the vertical stripes with gradual change.
Fig. 8 includes the figure being shown through shooting the value of G1sig and G2sig in the picture signal that subject is obtained.
Fig. 9 is the figure illustrating circular zone plate (CZP).
Figure 10 includes the value of G1sig and G2sig after being shown at the interpolation from the axle that the central horizontal of CZP extends
Figure.
Figure 11 is to illustrate the process selecting colour difference signal that selection circuit 320 according to a first embodiment of the present invention is carried out
Flow chart.
Figure 12 is the block diagram of the structure illustrating colour difference signal generative circuit 104 according to a second embodiment of the present invention.
Figure 13 is to illustrate the process selecting colour difference signal that selection circuit 320 according to a second embodiment of the present invention is carried out
Flow chart.
Figure 14 is the block diagram of the structure illustrating colour difference signal generative circuit 104 according to a third embodiment of the present invention.
Figure 15 is the process of the combination color difference signal illustrating that combiner circuit 380 according to a third embodiment of the present invention carried out
Flow chart.
Figure 16 is the figure illustrating bayer-pattern.
Detailed description of the invention
Embodiments of the invention are described in detail below with reference to accompanying drawing.
First embodiment
Fig. 1 is the block diagram of the structure illustrating image processing equipment according to a first embodiment of the present invention.
Image unit 101 includes unshowned pick-up lens, camera head and drive circuit thereof, and uses camera head
The optical imagery formed by pick-up lens is converted into the signal of telecommunication.Camera head includes charge coupled device (CCD) or complementary gold
Belong to oxide semiconductor (CMOS) sensor, and include the collection of pixels of the bayer-pattern shown in Figure 16.
By A/D converter 102, the analogue signal exported from image unit 101 is converted into digital signal.White balance
(WB) circuit 103 carries out known blank level adjustment to the picture signal being converted into digital signal by A/D converter 102.
The picture signal input exported from white balance circuit 103 is detected electricity to colour difference signal generative circuit 104, degree of association
Road 110 and luminance signal generative circuit 111.Degree of association testing circuit 110 is according to picture signal, inspection in units of pixel or region
The degree of association of survey signal in the vertical direction and the horizontal direction.Utilize the output of degree of association testing circuit 110, luminance signal raw
Become circuit 111 and brightness gamma circuitry 112 that the picture signal that white balance circuit 103 exports is carried out luminance signal process, thus
Luminance signal Y is exported in units of pixel or region.
Colour difference signal generative circuit 104 according to the picture signal inputted, in units of pixel or region generate aberration letter
Number R-G and B-G, and export colour difference signal.By color conversion matrix circuit 105, colour difference signal generative circuit 104 is exported
Colour difference signal R-G and B-G is converted into colour difference signal R-Y and B-Y respectively.Afterwards, color suppression circuit 106 is utilized to suppress high brightness
Color-difference gain with low-light level.Output signal from color suppression circuit 106 by known color gamma circuitry 107 and is used
In the known colourity flex point circuit 108 of adjustment cHroma gain, thus in units of pixel or region, export colour difference signal UV.
Will now describe the brightness interpolating process that luminance signal generative circuit 111 is carried out.Fig. 2 illustrate in camera head by
Subregion with 3 × 3 pixel compositions that the bayer-pattern shown in Figure 16 is constituted.This brightness interpolating process in, do not differentiate between from
The signal G1sig of the pixel output corresponding with G1 wave filter and the signal from the pixel output corresponding with G2 wave filter
G2sig, and using the two signal all as Gsig process.
First, luminance signal generative circuit 111 Gsig is interpolated into except G wave filter (G1 wave filter and G2 wave filter) with
The corresponding all pixels of outer wave filter, i.e. corresponding with R wave filter and B wave filter pixel.Such as, the P1 in Fig. 2～
P9 is the reference for identifying pixel, and R, G1, G2 represent the color filter corresponding with each pixel P1～P9 with B
Device." P5 (Rsig) " represents the value of the Rsig in pixel P5, and " P5 (Gsig) " represents the value of the Gsig in pixel P5, and " P5
(Bsig) " the value of the Bsig in expression pixel P5.
In the case of Gsig is interpolated into pixel P5 corresponding with B wave filter, detection is vertically and horizontally gone up
The degree of association of signal is to judge the direction of higher degree of relation.I.e., it is thus achieved that be positioned at the pixel of the pixel upper and lower sides as interpolation target
Signal between the absolute value of difference and difference between the signal of the pixel of the pixel left and right sides as interpolation target exhausted
To value.
Mathematical expression 1
Hdiff=| P4 (Gsig)-P6 (Gsig) | (1)
Vdiff=| P2 (Gsig)-P8 (Gsig) | (2)
Then, using the difference between Hdiff and Vdiff as interpolation walking direction signal MatSw.
Mathematical expression 2
MatSw=Hdiff-Vdiff (3)
When MatSw is for time negative, and the degree of association in horizontal direction is higher, thus by using the picture element signal in horizontal direction
Carry out interpolation.That is, using the meansigma methods of the Gsig of pixel P4 and the Gsig of pixel P6 as the Gsig of P5.On the other hand, be on duty into
Timing, by using the picture element signal in vertical direction to carry out interpolation.That is, flat by the Gsig of the Gsig of pixel P2 and pixel P8
Average is as the Gsig of P5.In a similar fashion all pixels corresponding with R wave filter and B wave filter are carried out inserting of Gsig
Value.
Then, Rsig is carried out by Gsig pair of all pixels corresponding with G wave filter and B wave filter after using interpolation
Interpolation.By using the Rsig of the pixel corresponding with the R wave filter being positioned at left and right sides or upper and lower sides and to these pixels
Carry out the Gsig that interpolation obtained and the pixel (P2, P4, P6 and P8) corresponding with G wave filter is carried out the interpolation of Rsig.And
And, by using the Rsig of the pixel corresponding with the R wave filter being positioned at around pixel and these pixels being carried out interpolation institute
The Gsig obtained, carries out interpolation according to the pixel that following formula pair is corresponding with B wave filter.
Mathematical expression 3
P2 (Rsig)=((P1 (Rsig)-P1 (Gsig))+(P3 (Rsig)-P3 (Gsig)))/2+P2 (Gsig)
(4)
P4 (Rsig)=((P1 (Rsig)-P1 (Gsig))+(P7 (Rsig)-P7 (Gsig)))/2+P4 (Gsig)
(5)
P5 (Rsig)=((P1 (Rsig)-P1 (Gsig))+(P3 (Rsig)-P3 (Gsig))+(P7 (Rsig)-P7
(Gsig))+(P9(Rsig)-P9(Gsig)))/4+P5(Gsig)···(6)
It is also possible to all pixels corresponding with G wave filter and R wave filter to be carried out the interpolation of Bsig.Therefore, respectively
Pixel obtains Rsig, Gsig and Bsig.Additionally, by using following formula (7) to calculate in each pixel, can obtain
Obtain luminance signal Y.
Mathematical expression 4
Y=0.3 × Rsig+0.59 × Gsig+0.11 × Bsig (7)
Note, can change as required expression formula (7) every in coefficient.
Then, the color interpolation carried out in explanation colour difference signal generative circuit 104 is processed.Fig. 3 is to illustrate colour difference signal
The block diagram of the structure of generative circuit 104.Colour difference signal generative circuit 104 includes color interpolation circuit 300 and 350, low-pass filtering
Device (LPF) 301～304 and 351～353, colour difference signal counting circuit 311～314, selection circuit 320 and false colour judge electricity
Road 360.
The picture signal exported by white balance circuit 103 inputs to color interpolation circuit 300.Fig. 4 is for color is described
The figure of the interpolation processing that interpolation circuit 300 is carried out.As shown in Figure 4, the figure that color interpolation circuit 300 will input with bayer-pattern
Image signal resolves into the picture signal being made up of the signal corresponding with each color of R, G1, G2 and B wave filter.Now, will letter
Number value 0 insertion not with decompose after the corresponding pixel of color filter in.
Afterwards, color interpolation circuit 500 uses known digital filter to utilize LPF to carry out interpolation processing, and exports institute
Pixel is had all to have color R signal, each, the picture signal of G1, G2 and B.
The picture signal of each color exported from color interpolation circuit 300 is inputted to having different cut-off frequency
LPF301～304.Fig. 6 (a) illustrates the example of the frequency characteristic of LPF301～304.It is special that LPF301 has the frequency in Fig. 6 (a)
Property 601, LPF302 has the frequency characteristic 602 in Fig. 6 (a), and LPF303 has the frequency characteristic 603 in Fig. 6 (a), and
LPF304 has the frequency characteristic 604 in Fig. 6 (a).From Fig. 6 (a) be appreciated that cut-off frequency with LPF301, LPF302,
The order of LPF303 and LPF304 increases.That is, the image with picture signal S01 exported from LPF301, from LPF302 output is believed
Number S02, picture signal S03 from LPF303 output and the order from picture signal S04 of LPF304 output comprise in the picture
Higher frequency content.Each picture signal S01～S04 have the image letter limiting each color R of frequency content, G1, G2 and B
Number.
By picture signal S01 input to colour difference signal counting circuit 311, picture signal S02 is inputted to colour difference signal meter
Calculate circuit 312, picture signal S03 is inputted to colour difference signal counting circuit 313, and the input of picture signal S04 is believed to aberration
Number counting circuit 314.Colour difference signal counting circuit 311～314 is by making in aforementioned manners to picture signal S01～S04 calculating
Colour difference signal.
That is, the object pixel that colour difference signal counting circuit 311 is obtained with reference to degree of association testing circuit 110 is in the horizontal direction
With the degree of association in vertical direction, thus generate colour difference signal R-G and B-G.Specifically, if degree of association testing circuit 110 is sentenced
Break as the degree of association in horizontal direction higher than the degree of association in vertical direction, then colour difference signal counting circuit 311 is by Rsig-
The value of G1sig as colour difference signal R-G and using the value of Bsig-G2sig as colour difference signal B-G.On the other hand, if degree of association
The degree of association that testing circuit 110 is judged as in vertical direction is higher than the degree of association in horizontal direction, then colour difference signal counting circuit
311 using the value of Rsig-G2sig as colour difference signal R-G and using the value of Bsig-G1sig as colour difference signal B-G.
Alternatively, colour difference signal counting circuit 311 by use the value with Rsig-G1sig and Rsig-G2sig value it
Between the corresponding weight coefficient of difference, using the meansigma methods of the value of Rsig-G1sig and the value of Rsig-G2sig as colour difference signal R-
G.Equally, colour difference signal counting circuit 311 is by using and the poor phase between the value of Bsig-G1sig and the value of Bsig-G2sig
Corresponding weight coefficient, using the meansigma methods of the value of Bsig-G1sig and the value of Bsig-G2sig as colour difference signal B-G.
Also in colour difference signal counting circuit 312～314, carry out above-mentioned process.
Colour difference signal R-G and B-G that colour difference signal counting circuit 311～314 is calculated as colour difference signal S11,
S12, S13 and S14 input is to selection circuit 320 described later.
Then explanation false colour is judged operation.The view data exported from white balance circuit 103 is inputted to color interpolation
Circuit 350.As color interpolation circuit 300, color interpolation circuit 350 exports all pixels and all has color signal, each
The picture signal of R, G1, G2 and B.
In the bayer-pattern shown in Figure 16, the pixel count corresponding with G wave filter is the pixel corresponding with R wave filter
The twice of number or the twice of the pixel count corresponding with B wave filter.By G filter equalizer becomes G1 wave filter and G2 filter
Ripple device, the pixel separation (sampling interval) of R, G1, G2 and B wave filter becomes the most equal.Therefore, by using the figure of G1 and G2
Image signal and detect aliasing noise occur region, it is possible to specify the picture signal of R and B occurs the region of aliasing noise.
The picture signal of each color exported from color interpolation circuit 350 is inputted to having different cut-off frequency
LPF351～353.Fig. 6 (b) illustrates the example of the frequency characteristic of LPF351～353.It is special that LPF351 has the frequency in Fig. 6 (b)
Property 611, LPF352 has the frequency characteristic 612 in Fig. 6 (b), and LPF353 has the frequency characteristic 613 in Fig. 6 (b).From
Fig. 6 (b) is appreciated that cut-off frequency increases with the order of LPF351, LPF352 and LPF353.That is, to export from LPF351
The order of picture signal T01, picture signal T02 from LPF352 output and picture signal T03 from LPF353 output is at image
In comprise higher frequency content.In the present embodiment, the frequency characteristic of LPF301 is corresponding with the frequency characteristic of LPF351,
The frequency characteristic of LPF302 is corresponding with the frequency characteristic of LPF352, and the frequency characteristic of LPF303 is special with the frequency of LPF353
Property is corresponding.
To input to false colour decision circuitry 360 from picture signal T01～the T03 of LPF351～353 output.
The method judging the region of pseudo-chrominance signal by false colour decision circuitry 360 referring now to Fig. 7 and 8 explanation.
Fig. 7 illustrates the subject of the vertical stripes with gradual change.Fig. 8 is shown through shooting what the image of the subject in Fig. 7 was obtained
The value of G1sig and G2sig in picture signal.
In Fig. 8 (a)～(i), transverse axis represents the position of the pixel in horizontal direction.In Fig. 8 (a), (d) and (g), and
The row G1sig that obtains from the pixel corresponding with the G1 wave filter particular row of configuration and from the G2 wave filter phase another row
The G2sig that corresponding pixel obtains.In the horizontal direction and the vertical direction, the sampling interval of corresponding with G1 wave filter pixel
With the sampling interval of the pixel corresponding with G2 wave filter is the most equal, and the pixel corresponding with G1 wave filter and with G2 filter
The space phase of the pixel that ripple device is corresponding offset by the half in sampling interval.
Owing to the subject in Fig. 7 is vertical stripes, thus subject image is incident to all row equably.Thus, when
When position according to the pixel in horizontal direction configures G1sig and G2sig, it is possible to obtain represent the subject shown in Fig. 7 gradually
The picture signal become.Here, in Fig. 8 (a), it is assumed that the frequency of the vertical stripes shown in Fig. 7 is sufficiently below G1 wave filter and G2
The nyquist frequency of wave filter.In Fig. 8 (d), it is assumed that the frequency of the vertical stripes shown in Fig. 7 is higher than hanging down shown in Fig. 8 (a)
The frequency of vertical bar stricture of vagina slightly less than G1 wave filter and the nyquist frequency of G2 wave filter.In Fig. 8 (g), it is assumed that shown in Fig. 7
The frequency of vertical stripes higher than G1 wave filter and the nyquist frequency of G2 wave filter.
In Fig. 8 (b), (e) and (h), only extract the G1sig in Fig. 8 (a), (d) and (g), and by using around
G1sig carries out the interpolation of G1sig to not corresponding with G1 wave filter pixel.In Fig. 8 (c), (f) and (i), only extract Fig. 8
(a), (d) and the G2sig in (g), and by using G2sig around that not corresponding with G2 wave filter pixel is carried out
The interpolation of G2sig.
It is appreciated that when the spatial frequency of subject is sufficiently below G1 wave filter and G2 wave filter from Fig. 8 (a)～(i)
During nyquist frequency, the phase offset of G1sig and G2sig after interpolation is little.Along with the spatial frequency of subject is closer to G1
Wave filter and the nyquist frequency of G2 wave filter, the phase offset of G1sig and G2sig after interpolation is big.When spatial frequency surpasses
When crossing nyquist frequency, the phase place of G1sig and G2sig after interpolation is substantially opposite.I.e., it is possible to after by detection interpolation
The phase contrast of G1sig and G2sig judges whether region is high-frequency region, i.e. whether region is the region that false colour occurs.
The operation of false colour decision circuitry 360 is described referring now to Fig. 9 and Figure 10.
Fig. 9 is the figure illustrating circular zone plate (CZP), wherein, using picture centre as the many concentric circulars of initial point configuration, and
And spatial frequency increases laterally from center.Figure 10 illustrates that is drawn is positioned at inserting from the axle of the central horizontal extension of CZP
The value of G1sig and G2sig after value.
In Figure 10 (a), curve 1001 represents the G1sig after the interpolation of picture signal T01, and curve 1002 represents image
G2sig after the interpolation of signal T01.In Figure 10 (b), curve 1003 represents the G1sig after the interpolation of picture signal T02, with
And curve 1004 represents the G2sig after the interpolation of picture signal T02.In Figure 10 (c), curve 1005 represents picture signal T03
Interpolation after G1sig, and curve 1006 represents the G2sig after the interpolation of picture signal T03.F0～F4 represents the sky of CZP
Between frequency.Spatial frequency at picture centre is represented by F0, and spatial frequency with the order of F1, F2, F3 and F4 outside image
Side is gradually increased.
False colour decision circuitry 360 is believed by the image of the G2sig after using the picture signal of the G1sig after interpolation and interpolation
The difference of the characteristic between number judges that whether region is the high-frequency region that pseudo-chrominance signal occurs.Specifically, after paying close attention to interpolation
The phase place of the slope of the picture signal of the G2sig after the picture signal of G1sig and interpolation, and when meeting following formula (8),
It is that the false colour region of pseudo-chrominance signal occurs by region decision.
Mathematical expression 5
Δ G1h × Δ G2h < 0 or Δ G1v × Δ G2v < 0 (8)
Here, Δ G1h represents the picture signal of the G1sig after interpolation slope in the horizontal direction, and Δ G2h represents
The picture signal of the G2sig after interpolation slope in the horizontal direction.And, Δ G1v represents the image letter of the G1sig after interpolation
Slope number in vertical direction, and Δ G2v represents the picture signal of the G2sig after interpolation slope in vertical direction.
For example, it is possible to by using the known digital filter shown in Fig. 5 (a) to obtain the slope in horizontal direction, and pass through
The known digital filter shown in Fig. 5 (b) is used to obtain the slope in vertical direction.For obtaining the wave filter of slope not
It is limited to this.
It is alternatively possible in units of the target area of each freedom multiple pixel composition, and do not sentence in units of pixel
Disconnected false colour region.In the case of judging in units of target area, if by each picture using this region to include
In the slope of picture signal that element is obtained, meet the percentage ratio of slope of the picture signal of expression formula (8) equal to or higher than threshold
Value, then can be false colour region by this region decision.Hereinafter, " target area " can be made up of multiple pixels region or
The region being made up of single pixel.
In the present embodiment, may determine that from Figure 10 (a), in picture signal T01, the spatial frequency in Fig. 9 is higher than
The region 2～4 of F1 is false colour region.And, may determine that from Figure 10 (b), the space frequency in picture signal T02, in Fig. 9
The rate region 3 and 4 higher than F2 is false colour region.Additionally, may determine that from Figure 10 (c), in picture signal T03, in Fig. 9
The spatial frequency region 4 higher than F3 is false colour region.
Selection circuit 320 is not the region 1 in false colour region for being judged as in picture signal T01, select from image
Signal T01 has the colour difference signal S11 that the picture signal of same frequency is generated.And, selection circuit 320 is for picture signal
Being judged as in T01 in false colour region and picture signal T02 being judged as is not the region 2 in false colour region, select from figure
Image signal T02 has the colour difference signal S12 that the picture signal of same frequency is generated.And, selection circuit 320 is believed for image
Being judged as in false colour region and picture signal T03 being judged as in number T02 is not the region 3 in false colour region, select from
Picture signal T03 has the colour difference signal S13 that the picture signal of same frequency is generated.And, selection circuit 320 is for image
Signal T03 is judged as the region 4 in false colour region, selects from the picture signal institute with the frequency less than picture signal T03
The colour difference signal S14 generated.By this way, selection circuit 320 can select not comprise the aberration of false colour in units of pixel
Signal.
In this embodiment it is assumed that the frequency characteristic of LPF301 is corresponding with the frequency characteristic of LPF351, the frequency of LPF302
Rate characteristic is corresponding with the frequency characteristic of LPF352, and the frequency characteristic of LPF303 is corresponding with the frequency characteristic of LPF353,
But the invention is not restricted to this.
Such as, it is judged as not being the region in false colour region in picture signal T01 to increase in picture signal S01
In there is not the probability of false colour, the upper limit of the frequency that can picture signal S01 be included is decreased to picture signal T01 and includes
Frequency the upper limit below.That is, the cut-off frequency of LPF351 is set to the cut-off frequency of slightly higher than LPF301.Equally, will
The cut-off frequency of LPF352 is set above the cut-off frequency of LPF302, and is set to slightly higher than by the cut-off frequency of LPF353
The cut-off frequency of LPF303.But, in this case, with the frequency characteristic of LPF301 corresponding to LPF351 frequency characteristic,
The frequency characteristic of LPF302 corresponds to the frequency of LPF353 corresponding to the frequency characteristic of LPF352 and the frequency characteristic of LPF303
The situation of characteristic is compared, and may increase the region of the color resolution with reduction.
On the contrary, when to reduce the region of the color resolution with reduction, can be by LPF301, LPF302 and LPF303
Cut-off frequency be set to the cut-off frequency of slightly less than corresponding LPF351, LPF352 and LPF353.But, in this situation
Under, picture signal S01 may comprise the frequency that the frequency than picture signal T01 is high.Thus, it is likely that by picture signal T01
In be judged as be not the region decision in false colour region be the false colour region in picture signal S01.
Suitably can change according to image to be generated whether make the cut-off frequency of LPF301～303 and LPF351～
The cut-off frequency of 353 is corresponding or arranges difference the most between which.Alternatively, because reducing the effect of resolution
Change according to the pattern of subject, it is possible to suitably change the cut-off frequency of these LPF according to subject.
As it has been described above, according to the present embodiment, colour difference signal generative circuit 104 generates by having different cut-off frequency
Multiple picture signals S01 of LPF301～304～S04, and hierarchically generate color according to multiple picture signals S01～S04
Difference signal S11～S14.Then, colour difference signal generative circuit 104 generate by have different cut-off frequency LPF351～
Multiple picture signals T01 of 353～T03, and in units of the picture signal of different layers, detect false colour region.Selection circuit 320
Picture signal according to which resolution from picture signal T01～T03 detects that false colour region selects colour difference signal S11
～any one in S14.Thus, selection circuit 320 can in units of pixel or region, select from the scope preventing false colour
The colour difference signal that picture signal in the layer of highest frequency is generated.
By this way, by amount necessary for color resolution reduction only to suppress false colour in the region that false colour occurs, and
And in the region that false colour does not occurs, elimination reduces the necessity of color resolution, such that it is able to realize the increase of color resolution
Suppression with false colour well balanced.
Figure 11 is the flow chart of the process selecting colour difference signal illustrating that selection circuit 320 carried out.Selection circuit 320 with
Pixel or region are that unit carries out selection process.It is assumed here that selection circuit 320 carries out selection process in units of pixel.
In step S1101, whether the pixel of picture signal T01 that selection circuit 320 judgement is corresponding with object pixel
It is judged as false colour region by false colour decision circuitry 360.If object pixel is judged as not being false colour region, then selection circuit
320 enter step S1104, and in step S1104, selection circuit 320 selects the aberration letter of the pixel corresponding with object pixel
Number S11 is as the colour difference signal of object pixel.
If object pixel is judged as being false colour region in step S1101, then selection circuit 320 enters step
S1102.In step S1102, selection circuit 320 judge the pixel of picture signal T02 corresponding with object pixel whether by
False colour decision circuitry 360 is judged as false colour region.If object pixel is judged as not being false colour region, then selection circuit 320
Entering step S1105, in step S1105, selection circuit 320 selects the colour difference signal of the pixel corresponding with object pixel
S12 is as the colour difference signal of object pixel.
If object pixel is judged as being false colour region in step S1102, then selection circuit 320 enters step
S1103.In step S1103, selection circuit 320 judge the pixel of picture signal T03 corresponding with object pixel whether by
False colour decision circuitry 360 is judged as false colour region.If object pixel is judged as not being false colour region, then selection circuit 320
Entering step S1106, in step S1106, selection circuit 320 selects the colour difference signal of the pixel corresponding with object pixel
S13 is as the colour difference signal of object pixel.
If object pixel is judged as being false colour region in step S1103, then selection circuit 320 enters step
S1107, in step S1107, selection circuit 320 selects the colour difference signal S14 of the pixel corresponding with object pixel as mesh
The colour difference signal of mark pixel.
Then, selection circuit 320 carries out colour difference signal selection process to all pixels of picture signal to be exported, and defeated
Go out for colour difference signal R-G and B-G comprised in any one in colour difference signal S11～S14 selected by each pixel.Then,
Colour difference signal R-G and B-G exported is input to color conversion matrix circuit 105 and is converted into signal R-Y and B-Y.
In the present embodiment, use expression formula (8) to judge whether region is false colour region, but the invention is not restricted to this.
For example, it is possible to pay close attention to Δ G1h, Δ G1v, Δ G2h and the value of Δ G2v, and if met in following formula (9)～(12)
Any one, then can be false colour region by region decision.And, threshold value TH in these expression formulas can be in picture signal T01
～it is mutually different in T03.Utilize the adjustment of these values, thus it is possible to vary for judging the benchmark in false colour region, and can adjust
Balance between increase and the suppression of false colour of color resolution.
Mathematical expression 6
Δ G1h-TH>0 and Δ G2h+TH<0 (9)
Δ G1h+TH<0 and Δ G2h-TH>0 (10)
Δ G1v-TH>0 and Δ G2v+TH<0 (11)
Δ G1v+TH<0 and Δ G2v-TH>0 (12)
(wherein, TH > 0)
Pay close attention to the slope in horizontal direction and the slope in vertical direction to judge whether region is false colour region.Optional
Ground, can pay close attention to the slope on incline direction.
And, the slope of the picture signal of the G2sig after the slope of the picture signal of the G1sig after use interpolation and interpolation
Judge whether region is false colour region.It is alternatively possible to use the difference between G1sig and G2sig.If that is, G1sig and
Difference between G2sig is big, then may determine that as phase offset big.Specifically, even if being unsatisfactory in expression formula (8)～(12)
Any one, if as shown in expression formula (13), the absolute value of the difference of G1sig and G2sig in object pixel exceedes threshold value, it is possible to
Object pixel to be judged as false colour region.
Mathematical expression 7
|G1sig-G2sig|>0···(13)
But, if luminance signal is little, then the value of G1sig and G2sig is the least.Thus, it is possible to generate according to luminance signal
The luminance signal of the corresponding pixel that circuit 111 is generated is to change these threshold values.
Alternatively, except the picture signal of the G2sig after the slope of picture signal based on the G1sig after interpolation and interpolation
Slope judgement beyond, it is also possible to combine the judgement of difference based on the value between G1sig and G2sig.
Additionally, these Rule of judgment can change between picture signal T01～T03.
Additionally, in addition to object pixel, it is also possible to by considering sentencing of the false colour region of the pixel around object pixel
Disconnected result judges whether object pixel is false colour region.
In the present embodiment, generate colour difference signal according to four layer image signals, and in three layer image signals, carry out puppet
The judgement in territory, zone, but the number of plies is not limited to this.Colour difference signal can be generated from the picture signal of the different frequency of N+1 layer, and
And the judgement in false colour region can be carried out in the picture signal of the different frequency of N shell.
In the present embodiment, color interpolation circuit 300 and color interpolation circuit 350 are separated from each other, but if they tools
There is common characteristic, then can use a common circuit.It is also possible to have altogether at LPF301～303 and LPF351～353
With using a common circuit as LPF301～303 and LPF351～353 in the case of characteristic.
In the present embodiment, illustrate to select the colour difference signal that the picture signal of the different frequency from multiple layers generates
Any one is to suppress the example of false colour, but the invention is not restricted to this.In method same as described above, can from N+1 layer not
The picture signal of same frequency generates luminance signal, and can for the picture signal of the different frequency of N shell in the same way
Detection moire fringes region, and do not detect false colour region, such that it is able to the moire fringes that suppression causes due to luminance signal.That is, this reality
It is effective for executing example for the moire fringes caused due to brightness and the false colour caused due to color signal.If additionally, to carry out
The picture signal processed for reducing the size adjustment of pixel count carries out the process according to the present embodiment, then can suppress size
Moire fringes caused by adjustment process or false colour.
As it has been described above, the image letter obtained from the camera head of bayer-pattern according to the image processing equipment of the present embodiment
Extract the picture signal of the G1 wave filter being used as the first color filter in number and be used as the G2 wave filter of the second color filter
Picture signal, row interpolation of going forward side by side.With the pixel that identical period assignment is corresponding with G1 wave filter and G2 wave filter, and with
The space phase of the pixel that G1 wave filter is corresponding with G2 wave filter is shifted by.Thus, as shown in figures 8 and 10, at high-frequency region
The phase place of the picture signal of the G2 wave filter after the picture signal of the G1 wave filter after middle reversion interpolation and interpolation.Thus, image
Processing equipment can be based on G1 wave filter and the slope of the picture signal of G2 wave filter and G1 wave filter and the figure of G2 wave filter
At least one in both differences between image signal judges whether target area (object pixel) is high-frequency region.If it is determined that
Be high-frequency region for target area, then image processing equipment generates such as colour difference signal and brightness according to following picture signal
The prearranged signals of signal etc. such that it is able to the pseudo-normal complexion moire fringes of suppression, wherein, in this picture signal, cut-off frequency is reduced
The grade of high-frequency region it is not judged as to object pixel.
Alternatively, image processing equipment is so that the equipment of after-stage generates such as colour difference signal and luminance signal etc.
Prearranged signals, and the mark of the judged result representing that whether object pixel be high-frequency region can be set in picture signal.
Second embodiment
Second embodiment is when selection circuit 320 selects colour difference signal also to judge with the difference of first embodiment
Whether object pixel is marginal area.In the first embodiment, if false colour decision circuitry 360 is judged as that target area is false colour
Region, then selection circuit 320 selects the colour difference signal generated from the picture signal of lower frequency.As a result, if the puppet of being judged as
The region in territory, zone is marginal area, then the color in marginal area is it may appear that ooze out.
The structure of the image processing equipment according to the present embodiment and the structure of the image processing equipment according to first embodiment
Identical, and thus the description thereof will be omitted.In the present embodiment, of the structure of the colour difference signal generative circuit 104 shown in Fig. 1
Divide different from first embodiment.
Figure 12 is the block diagram of the structure illustrating the colour difference signal generative circuit 104 according to the present embodiment.Shown in Fig. 3
Beyond the structure of colour difference signal generative circuit, also include according to the colour difference signal generative circuit 104 of the present embodiment receiving from
Picture signal T01 of LPF351～353 output～the edge decision circuitry 370 of T03.Edge decision circuitry 370 is generated
The judged result relevant with marginal area inputs to selection circuit 320.
Edge decision circuitry 370 is by using expression formula (14), comprising based on from picture signal T01 that LPF351 exports
The picture signal of G1 and the picture signal of G2 calculate the Gsig in all pixels.
Mathematical expression 8
Gsig=(G1sig+G2sig)/2 (14)
Then, the picture signal of concern Gsig slope in the horizontal direction and the vertical direction.If meeting following expression
Formula (15), then be judged as object pixel marginal area, and will determine that result exports to selection circuit 320.
Mathematical expression 9
Δ Gh > TH or Δ Gv > TH (15)
Here, Δ Gh represents the picture signal of the Gsig after interpolation slope in the horizontal direction, and Δ Gv represents slotting
The picture signal of the G2sig after value slope in the horizontal direction.The known digital filter shown in Fig. 5 (a) can be passed through
And obtain these slopes by the known digital filter shown in Fig. 5 (b) in vertical direction.For obtaining slope
Wave filter is not limited to this.It is alternatively possible in units of the region being made up of multiple pixels, and do not judge in units of pixel
Marginal area.In the case of judging in units of region, if by using each pixel that this region includes to obtain
Picture signal slope in, meet the percentage ratio of slope of the picture signal of expression formula (15) equal to or higher than threshold value, then
Can be marginal area by this region decision.
Equally, the edge decision circuitry 370 picture signal T02 to exporting from LPF352 and the image letter from LPF353 output
Number T03 carries out the judgement of marginal area.
Having been described that uses the example of Gsig as the method for judging marginal area based on G1sig and G2sig.Can
Selection of land, it is also possible to use the luminance signal by using expression formula (7) to be generated from Rsig, Gsig and Bsig.
False colour decision circuitry 360 carries out the judgement in the false colour region in the process identical with first embodiment, and will determine that
Result exports to selection circuit 320.
Figure 13 is the flow process of the process selecting colour difference signal illustrating that the selection circuit 320 according to the present embodiment carried out
Figure.Selection circuit 320 carries out selection process in units of pixel or region.It is assumed here that selection circuit 320 is in units of pixel
Carry out selection process.
In step S1301, whether the pixel of picture signal T01 that selection circuit 320 judgement is corresponding with object pixel
It is judged as false colour region by false colour decision circuitry 360 and is judged as it not being marginal area by edge decision circuitry 370.If picture
Element is judged as it not being false colour region by false colour decision circuitry 360, then selection circuit 320 enters step S1304 to select and this mesh
The colour difference signal S11 of the pixel that mark pixel is corresponding is as the colour difference signal of object pixel.And, if pixel is judged by edge
Circuit 370 is judged as marginal area, then selection circuit 320 enters step S1304 to select the picture corresponding with this object pixel
The colour difference signal S11 of element is as the colour difference signal of object pixel.That is, even if object pixel is judged as false colour region, in target
In the case of pixel is judged as marginal area, selection circuit 320 also selects colour difference signal S11 with in preferential suppression marginal area
Oozing out of color.
If pixel is judged as false colour region by false colour decision circuitry 360 and is judged as not by edge decision circuitry 370
Be marginal area, then selection circuit 320 enters step S1302.In step S1302, selection circuit 320 judges and object pixel
Whether the pixel of corresponding picture signal T02 is judged as false colour region and by edge judgement electricity by false colour decision circuitry 360
Road 370 is judged as it not being marginal area.If pixel is judged as it not being false colour region by false colour decision circuitry 360, then select electricity
Road 320 enters step S1305 to select the colour difference signal S12 color as object pixel of the pixel corresponding with this object pixel
Difference signal.And, if pixel is judged as marginal area by edge decision circuitry 370, then selection circuit 320 enters step
S1305 is to select the colour difference signal S12 colour difference signal as object pixel of the pixel corresponding with this object pixel.I.e., i.e.
Making object pixel be judged as false colour region, in the case of object pixel is judged as marginal area, selection circuit 320 also selects
Select colour difference signal S12 with the oozing out of color in preferential suppression marginal area.
If pixel is judged as false colour region by false colour decision circuitry 360 and is judged as not by edge decision circuitry 370
Be marginal area, then selection circuit 320 enters step S1303.In step S1303, selection circuit 320 judges and object pixel
Whether the pixel of corresponding picture signal T03 is judged as false colour region and by edge judgement electricity by false colour decision circuitry 360
Road 370 is judged as it not being marginal area.If pixel is judged as it not being false colour region by false colour decision circuitry 360, then select electricity
Road 320 enters step S1306 to select the colour difference signal S13 color as object pixel of the pixel corresponding with this object pixel
Difference signal.And, if pixel is judged as marginal area by edge decision circuitry 370, then selection circuit 320 enters step
S1306 is to select the colour difference signal S13 colour difference signal as object pixel of the pixel corresponding with this object pixel.I.e., i.e.
Making object pixel be judged as false colour region, in the case of object pixel is judged as marginal area, selection circuit 320 also selects
Select colour difference signal S13 with the oozing out of color in preferential suppression marginal area.
On the contrary, if pixel is judged as false colour region by false colour decision circuitry 360 and is judged by edge decision circuitry 370
For not being marginal area, then selection circuit 320 enters step S1307 to select colour difference signal S14.
As it has been described above, in the present embodiment, even if region is judged as false colour region by false colour decision circuitry 360, in region
In the case of being marginal area, selection circuit 320 also selects the frequency that the frequency according to the situation than region not being marginal area is high
The colour difference signal that picture signal in the layer of rate is generated.
In this way it is possible to obtain the effect obtained in the structure according to first embodiment, and also can suppress
In marginal area, color oozes out.
3rd embodiment
With the difference of first embodiment, 3rd embodiment is that colour difference signal generative circuit 104 is not output aberration letter
Any one in number S11～S14, but output is by synthesizing the colour difference signal that these colour difference signals S11～S14 is generated.
The structure of the image processing equipment according to the present embodiment and the structure of the image processing equipment according to first embodiment
Identical, and thus the description thereof will be omitted.In the present embodiment, of the structure of the colour difference signal generative circuit 104 shown in Fig. 1
Divide different from first embodiment.
Figure 14 is the block diagram of the structure illustrating the colour difference signal generative circuit 104 according to the present embodiment.According to the present embodiment
Colour difference signal generative circuit 104 include that combiner circuit 380 is to replace the selection circuit 320 shown in Fig. 3.
Figure 15 is the flow process of the process of the combination color difference signal illustrating that the combiner circuit 380 according to the present embodiment carried out
Figure.Combiner circuit 380 carries out selection process in units of pixel or region.It is assumed here that combiner circuit 380 is in units of pixel
Carry out selection process.Combiner circuit 380 is according to pseudo-zone being detected in which picture signal in picture signal T01～T03
Territory changes the weight of colour difference signal S11～S14, combination color difference signal S11～S14, and exports generated signal.
In step S1501, whether the pixel of picture signal T01 that combiner circuit 380 judgement is corresponding with object pixel
It is judged as false colour region by false colour decision circuitry 360.If pixel is judged as it not being false colour region by false colour decision circuitry 360,
Then combiner circuit 380 enters step S1504 and is set to the weight by the colour difference signal S11 of the pixel corresponding with object pixel
Weight limit relative to the weight of other colour difference signal.
If pixel is judged as false colour region by false colour decision circuitry 360 in step S1501, then combiner circuit 380 enters
Enter step S1502.In step S1502, combiner circuit 380 judges the pixel of picture signal T02 corresponding with object pixel
Whether it is judged as false colour region by false colour decision circuitry 360.If pixel is judged as it not being pseudo-zone by false colour decision circuitry 360
Territory, then combiner circuit 380 enters step S1505 and sets with the weight by the colour difference signal S12 of the pixel corresponding with object pixel
It is set to the weight limit of the weight relative to other colour difference signal.
If pixel is judged as false colour region by false colour decision circuitry 360 in step S1502, then combiner circuit 380 enters
Enter step S1503.In step S1503, combiner circuit 380 judges the pixel of picture signal T03 corresponding with object pixel
Whether it is judged as false colour region by false colour decision circuitry 360.If pixel is judged as it not being pseudo-zone by false colour decision circuitry 360
Territory, then combiner circuit 380 enters step S1506 and sets with the weight by the colour difference signal S13 of the pixel corresponding with object pixel
It is set to the weight limit of the weight relative to other colour difference signal.
On the contrary, if pixel is judged as false colour region by false colour decision circuitry 360, then combiner circuit 380 enters step
S1507 is arranged with respect to other colour difference signal with the weight by the colour difference signal S14 of the pixel corresponding with object pixel
The weight limit of weight.
Combiner circuit 380 is according to the weight arranged in step S1504～S1507, by using following arbitrarily expression formula
Combination color difference signal S11～S14.
Mathematical expression 10
(R-G, B-G)=(6 × S11+2 × S12+S13+S14)/10 (16)
(R-G, B-G)=(2 × S11+5 × S12+2 × S13+S14)/10 (17)
(R-G, B-G)=(S11+2 × S12+5 × S13+2 × S14)/10 (18)
(R-G, B-G)=(S11+S12+2 × S13+6 × S14)/10 (19)
If the weight of colour difference signal S11 is arranged to maximum, then combiner circuit 380 selects expression formula (16).Equally, as
Really the weight of colour difference signal S12, S13 and S14 is arranged to maximum, then combiner circuit 380 selects expression formula (17) respectively, expresses
Formula (18) and expression formula (19).Then, combiner circuit 380 is defeated by processing, by this synthesis, colour difference signal R-G and B-G obtained
Go out to color conversion matrix circuit 105.
As in the second embodiment, it may be considered that whether object pixel is marginal area.Even if that is, object pixel quilt
It is judged as false colour region, in the case of object pixel is judged as marginal area, it is also possible to will be from being judged as marginal area
Picture signal in there is the weight of the colour difference signal that the picture signal of higher cutoff frequency is generated be set to maximum.
It addition, combiner circuit 380 can change aberration letter according to the pseudo-colourity detected from picture signal T01～T03
The weight of number S11～S14.The Δ G1h that can obtain according to the G1sig after interpolation and the G2sig after interpolation, Δ G1v, Δ
Difference between G2h and Δ G2v and G1sig and the value of G2sig obtains pseudo-colourity.That is, combiner circuit 380 arranges pseudo-colourity,
So that the phase offset of the G1sig after interpolation and the G2sig after interpolation is estimated as closer to 180 degree, the value of pseudo-colourity is more
Greatly.And, combiner circuit 380 can arrange the power of colour difference signal S11～S14 according to the pseudo-colourity of picture signal T01～T03
Weight.
Other embodiments
The purpose of the present invention is realized also by following methods.That is, will be stored for realizing function soft of above-described embodiment
Storage medium (or the record medium) supply of part program code is to system or equipment.Then, system or equipment computer (or
CPU or MPU etc.) read and perform the program code of storage in storage medium.In this case, the journey read from storage medium
Sequence code itself realizes the function of above-described embodiment, and the storage medium storing program code constitutes the present invention.And, except
When realizing the function of above-described embodiment when computer performs the program code read, situations below is also included within
In the present invention.Operating system (OS) instruction based on program code run i.e., in a computer carries out all or part of reality
Border processes, and is realized the function of above-described embodiment by process.
Additionally, situations below is included in the invention.That is, the program code read from storage medium writes on insertion to calculate
In function expansion card in machine arrange memorizer or be connected on the functional expansion unit of computer.Then, Function Extension
The instructions based on program code such as the CPU arranged in card or functional expansion unit carry out all or part of actual treatment, and
The function of above-described embodiment is realized by process.
In the case of the present invention is applied to above-mentioned storage medium, by the program code storage corresponding with said process
In storage medium.
Reference numerals list
101 image units
102 A/D converters
103 white balances (WB) circuit
104 colour difference signal generative circuits
105 color conversion matrix circuit
106 color suppression circuits
107 color gamma circuitry
108 colourity flex point circuit
110 degree of association testing circuits
111 luminance signal generative circuits
112 brightness gamma circuitry
300,350 color interpolation circuit
301,302,303,304,351,352,353 low pass filter (LPF)
311,312,313,314 colour difference signal counting circuit
320 selection circuits
360 false colour decision circuitry
370 edge decision circuitry
380 combiner circuits
Claims (12)
1. an image processing equipment, it is characterised in that including:
Extracting parts, for extracting multiple picture signals with different cut-off frequency from the picture signal of input；
Judging parts, at least one in the multiple picture signals extracted for described extracting parts judges that target area is
No is high-frequency region；And
Processing component, in the case of described judgement parts are not judged as that described target area is high-frequency region, exports institute
State in target area be judged as described target area be high-frequency region situation compared with, there is the image of higher frequency content
Signal.
Image processing equipment the most according to claim 1, it is characterised in that
The output of described processing component is not judged as that described target area is to have highest frequency in the picture signal of high-frequency region
The picture signal of composition.
Image processing equipment the most according to claim 1, it is characterised in that also include:
Detection part, is used for judging whether described target area is marginal area,
Wherein, the output of described processing component is judged as that described target area is to have high frequency in the picture signal of marginal area
The picture signal of rate composition.
Image processing equipment the most according to claim 1, it is characterised in that
The picture signal of described input obtains from camera head,
Wherein, in described camera head, it is configured with multiple color filter and being provided with in a predetermined pattern and filters with each color
The pixel that ripple device is corresponding.
Image processing equipment the most according to claim 4, it is characterised in that
Described camera head includes with red filter, blue electric-wave filter and the green filter of bayer-pattern configuration,
Described judgement parts use the pixel corresponding with the first green filter picture signal and with the second green filter
The picture signal of corresponding pixel judges whether described target area is high-frequency region, wherein, and described first green filter
It is configured in the horizontal direction of described red filter and is configured in the vertical direction of described blue electric-wave filter, and
Described second green filter is configured in the vertical direction of described red filter and is configured in described blue color filtered
In the horizontal direction of device.
6. an image processing equipment, it is characterised in that including:
Extracting parts, for extracting multiple picture signals with different cut-off frequency from the picture signal of input；
Judging parts, at least one in the multiple picture signals extracted for described extracting parts judges that target area is
No is high-frequency region；And
Processing component, in the case of described judgement parts are not judged as that described target area is high-frequency region, increase with
It is judged as that described target area is the weight that the picture signal of higher frequency content was compared, had to the situation of high-frequency region, and
And synthesize the plurality of picture signal so that the described target area of output to synthesize the signal obtained.
Image processing equipment the most according to claim 6, it is characterised in that
Described processing component is not by being judged as that described target area is to have high frequency in the picture signal of high-frequency region
The weight of the picture signal of rate composition is set greater than the weight of other picture signal, carries out described synthesis.
Image processing equipment the most according to claim 6, it is characterised in that also include:
Detection part, is used for judging whether described target area is marginal area,
Wherein, described processing component by will be deemed as in the picture signal that described target area is marginal area have the highest
The weight of the picture signal of frequency content is set greater than the weight of other picture signal, carries out described synthesis.
Image processing equipment the most according to claim 6, it is characterised in that
The picture signal of described input obtains from camera head,
Wherein, in described camera head, it is configured with multiple color filter and being provided with in a predetermined pattern and filters with each color
The pixel that ripple device is corresponding.
Image processing equipment the most according to claim 9, it is characterised in that
Described camera head includes with red filter, blue electric-wave filter and the green filter of bayer-pattern configuration,
Described judgement parts use the picture signal obtained from the first green filter and the figure obtained from the second green filter
Image signal judges whether described target area is high-frequency region, and wherein, described first green filter is configured in described red
In the horizontal direction of color filter and be configured in the vertical direction of described blue electric-wave filter, and the described second green filter
Ripple device is configured in the vertical direction of described red filter and is configured in the horizontal direction of described blue electric-wave filter.
11. 1 kinds of image processing methods, it is characterised in that including:
Extraction step, for extracting multiple picture signals with different cut-off frequency from the picture signal of input；
Judging step, at least one in the multiple picture signals extracted for described extraction step judges that target area is
No is high-frequency region；And
Process step, in the case of described judgement step is not judged as that described target area is high-frequency region, export institute
State in target area be judged as described target area be high-frequency region situation compared with, there is the image of higher frequency content
Signal.
12. 1 kinds of image processing methods, it is characterised in that including:
Extraction step, for extracting multiple picture signals with different cut-off frequency from the picture signal of input；
Judging step, at least one in the multiple picture signals extracted for described extraction step judges that target area is
No is high-frequency region；And
Process step, in the case of described judgement step is not judged as that described target area is high-frequency region, increase with
It is judged as that described target area is the weight that the picture signal of higher frequency content was compared, had to the situation of high-frequency region, and
And synthesize the plurality of picture signal so that the described target area of output to synthesize the signal obtained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410201465.3A CN104010174B (en) | 2009-10-13 | 2009-10-13 | Image processing equipment and image processing method |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410201465.3A CN104010174B (en) | 2009-10-13 | 2009-10-13 | Image processing equipment and image processing method |
| CN200980161949.7A CN102577394B (en) | 2009-10-13 | 2009-10-13 | Image processing device and image processing method |
Related Parent Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN200980161949.7A Division CN102577394B (en) | 2009-10-13 | 2009-10-13 | Image processing device and image processing method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104010174A CN104010174A (en) | 2014-08-27 |
| CN104010174B true CN104010174B (en) | 2016-08-31 |
Family
ID=51370649
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410201465.3A Active CN104010174B (en) | 2009-10-13 | 2009-10-13 | Image processing equipment and image processing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104010174B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107493407B (en) * | 2016-06-08 | 2021-08-31 | 深圳富泰宏精密工业有限公司 | Photographing device and photographing method |
| CN107786858B (en) * | 2016-08-25 | 2019-07-26 | 北京展讯高科通信技术有限公司 | A kind of method, apparatus and electronic equipment inhibiting false colour based on spatial frequency |
| CN106507066B (en) * | 2016-11-22 | 2018-04-17 | 深圳开阳电子股份有限公司 | Suppress the method, apparatus of false color of image |
| CN108615227B (en) * | 2018-05-08 | 2021-02-26 | 浙江大华技术股份有限公司 | Image moire inhibition method and device |
| CN108769523B (en) * | 2018-06-07 | 2020-06-30 | Oppo广东移动通信有限公司 | Image processing method and apparatus, electronic device, computer-readable storage medium |
| CN113905193B (en) * | 2020-06-18 | 2025-02-18 | 思特威(上海)电子科技股份有限公司 | Image sensor with dark scene full-color function and imaging method thereof |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1610412A (en) * | 2003-10-23 | 2005-04-27 | 索尼株式会社 | Image processing device, image processing method, and program |
| CN1957618A (en) * | 2004-05-26 | 2007-05-02 | 安奇逻辑股份有限公司 | Imaging device |
| JP2008072377A (en) * | 2006-09-13 | 2008-03-27 | Canon Inc | Luminance signal creating device, luminance signal creating method and image pickup device |
| WO2011045849A1 (en) * | 2009-10-13 | 2011-04-21 | キヤノン株式会社 | Imaging device |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5171361B2 (en) * | 2008-04-07 | 2013-03-27 | 株式会社日立製作所 | Imaging device |
-
2009
- 2009-10-13 CN CN201410201465.3A patent/CN104010174B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1610412A (en) * | 2003-10-23 | 2005-04-27 | 索尼株式会社 | Image processing device, image processing method, and program |
| CN1957618A (en) * | 2004-05-26 | 2007-05-02 | 安奇逻辑股份有限公司 | Imaging device |
| JP2008072377A (en) * | 2006-09-13 | 2008-03-27 | Canon Inc | Luminance signal creating device, luminance signal creating method and image pickup device |
| WO2011045849A1 (en) * | 2009-10-13 | 2011-04-21 | キヤノン株式会社 | Imaging device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104010174A (en) | 2014-08-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP5490139B2 (en) | Imaging device | |
| CN104010174B (en) | Image processing equipment and image processing method | |
| JP6492055B2 (en) | Device for acquiring bimodal images | |
| JP4284628B2 (en) | IMAGING DEVICE, IMAGE PROCESSING DEVICE, IMAGE PROCESSING METHOD, PROGRAM FOR IMAGE PROCESSING METHOD, AND RECORDING MEDIUM CONTAINING PROGRAM FOR IMAGE PROCESSING METHOD | |
| US7973831B2 (en) | Solid-state image sensor and imaging apparatus using the same | |
| JP5045421B2 (en) | Imaging apparatus, color noise reduction method, and color noise reduction program | |
| US8270774B2 (en) | Image processing device for performing interpolation | |
| US8229217B2 (en) | Image processing method and apparatus, image processing program and medium storing this program | |
| CN101494795B (en) | Image sensor and method using unit pixel groups with overlapping green spectral ranges | |
| CN103202022A (en) | Image processing apparatus and control method of the same | |
| KR102224851B1 (en) | Image Processing Device and Image Processing System performing sub-pixel interpolation | |
| US9143747B2 (en) | Color imaging element and imaging device | |
| US8471933B2 (en) | Image processing apparatus, image processing method and computer program | |
| CN101646090B (en) | Luminance signal generation apparatus, luminance signal generation method, and image capturing apparatus | |
| CN104471929B (en) | Color image sensor and camera head | |
| CN108122201A (en) | A kind of Bayer interpolation slide fastener effect minimizing technology | |
| US9185375B2 (en) | Color imaging element and imaging device | |
| KR20110130266A (en) | An image processing apparatus, an image processing method, and a recording medium storing a program for executing the same | |
| JP2012095341A (en) | Imaging apparatus | |
| JP5766274B2 (en) | Image processing apparatus and image processing method | |
| EP2001244B1 (en) | Signal processing method and signal processing device | |
| CN119496995A (en) | Image processing method, device and equipment | |
| KR20140133407A (en) | Syestem and method for image processing | |
| Li et al. | Comparison study on color filter array interpolating methods | |
| JP2018107682A (en) | Luminance signal generating device, luminance signal generating method, and program |
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 |