CN101808249B - Image processing circuit and image processing method - Google Patents
Image processing circuit and image processing method Download PDFInfo
- Publication number
- CN101808249B CN101808249B CN2009100090134A CN200910009013A CN101808249B CN 101808249 B CN101808249 B CN 101808249B CN 2009100090134 A CN2009100090134 A CN 2009100090134A CN 200910009013 A CN200910009013 A CN 200910009013A CN 101808249 B CN101808249 B CN 101808249B
- Authority
- CN
- China
- Prior art keywords
- image signal
- image
- pixel
- weights
- colourity
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Processing Of Color Television Signals (AREA)
Abstract
The invention discloses an image processing circuit and an image processing method. The image processing method can effectively deal with the phenomenon of abnormal color blocks produced by processing multi-pulse patterns in the prior art, can dynamically adjust the weight ratio so as to widely deal with the distribution situations of various colors in images, and is used for processing images based on a SECAM-system system. The image comprises a first pixel, a second pixel and a third pixel which are sequentially arranged on the same vertical line, and respectively correspond to a first video signal, a second video signal and a third video signal. The image processing method comprises the following steps: firstly, according to the first video signal, the second video signal and the thirdvideo signal, working out a chroma signal through a vertical filtering procedure; secondly, according to the chroma signal, working out a chroma angular frequency through a frequency demodulation procedure; and finally, according to the chroma angular frequency, producing a chroma value.
Description
Technical field
The present invention relates to a kind of image treatment method and image-processing circuit, refer to a kind of image treatment method and image-processing circuit especially based on SEQUAM system (SECAM) system.
Background technology
(composite video baseband signal is a kind of analog video signal that widely uses at present CVBS) to synthetic video signal baseband signal, and is that brightness (luma) signal and colourity (chroma) signal by image mixes.The related specifications of CVBS mainly contains three major types: NTSC (NationalTelevision System Committee; NTSC) system, line-by-line inversion (Phase Alternating Line; PAL) system and SEQUAM system (S é quential Couleur Avec M é moire, SECAM) system.
See also Fig. 1, Fig. 1 is the waveform sketch map of CVBS.Prompting signal 10 is in order to indicate the initial part of every horizontal scanning line synchronously.It is wavy that color burst (color burst) 12 presents a string, in order to the reference signal relevant with colourity to be provided.In Fig. 1, after color burst 12, presenting stair-stepping signal is the luminance signal of representing in the image signal.For instance, the v among Fig. 1
MAXPossibly be 255 brightness corresponding to the GTG value, v
MINThen possibly be 0 brightness corresponding to the GTG value.
In the standard of SECAM-system; Chrominance signal is to show by adding all string wave tables on luminance signal; As being positioned at the string ripple signal on the luminance signal among Fig. 1, borrowing the angular frequency that compares between chrominance signal and the color burst, just can calculate corresponding chromatic value (color).
Fig. 2 shows the image-processing circuit of the chromatic component that is used to decipher the Sequential Color and Memory system signal in the known techniques.Image-processing circuit 1 comprises receiving element 13, filter unit 11, frequency demodulation module 14 and chroma conversion unit 16.Filter unit 11 is coupled to receiving element 13, and frequency demodulation mould module 14 is coupled to filter unit 11, and chroma conversion unit 16 is coupled to frequency demodulation module 14.
Receiving element 13 is followed the coded image of SECAM-system standard in order to reception; The image signal of some pixels can be expressed as in the image: S=Y+Sin [(ω c+ Δ ω) * t+ φ 1]; Wherein Y is a luminance signal, and Sin [(ω c+ Δ ω) * t] is a chrominance signal, and ω c is the carrier angular frequencies of SECAM-system; Δ ω is the hunting angle frequency of image signal and the difference on the frequency of carrier angular frequencies, and φ 1 is a phase place.Because the chrominance signal of Sequential Color and Memory system is to encode with the frequency modulating mode, so Δ ω can a corresponding chromatic value.
In the SECAM-system, the chrominance signal of image is that carrier angular frequencies can be 4.25MHz and 4.41MHz with carrier angular frequencies (carrier frequency) change in oscillation.See also Fig. 3 (a); If the luminance signal among Fig. 3 (a) is with near the hunting of frequency of carrier angular frequencies the time; Then the luminance signal in this image signal just can't be with the image treatment method filtering of prior art, and reason is that the frequency demodulation (frequency demodulation) that the radio-frequency component of luminance signal also has been brought into chrominance signal is calculated.So in the frequency demodulation of image treatment method was calculated, the rule vibration of luminance signal had the chromatic value (color) of a correspondence, thereby in GTG multiple-pulse pattern, has produced unusual color lump.Fig. 3 (b) is the sketch map that produces unusual color lump in the multiple-pulse pattern, and rectangular being spaced of a plurality of black constitutes GTG multiple-pulse pattern, and the color lump 20 with color is formed in the multiple-pulse pattern.
Therefore, how avoiding the unusual color lump phenomenon in the above-mentioned multiple-pulse pattern takes place, is the subject matter that image treatment method of the present invention and image-processing circuit desire solve.
Summary of the invention
Technical problem to be solved by this invention provides a kind of image-processing circuit and image treatment method; Can solve effectively and handle the unusual color lump phenomenon that the multiple-pulse pattern is produced in the prior art; And dynamically adjust the weights ratio, to handle shades of colour distribution situation in the image widely.
In order to solve above technical problem, the invention provides following technical scheme:
The invention provides a kind of image treatment method,, and can solve the unusual color lump phenomenon that processing multiple-pulse pattern is produced in the prior art in order to the image of processing based on SECAM system.
According to a specific embodiment, this image comprises a plurality of pixels, and these pixels comprise one first pixel, one second pixel and one the 3rd pixel of continuous arrangement at least, and respectively corresponding one first image signal, one second image signal and one the 3rd image signal.This image treatment method comprises the following step.
At first, execution in step (a) according to this first image signal, this second image signal and the 3rd image signal, and via vertical filtering (vertical filter) program, calculates one first chrominance signal; Then, execution in step (b) according to this first chrominance signal, and via a frequency demodulation (frequencydemodulation) program, calculates one first colourity angular frequency; At last, execution in step (c) according to this first colourity angular frequency, calculates one first chromatic value.
In addition, image treatment method of the present invention can also calculate color correct in the image according to the different colours distribution situation of reception image.
At first, execution in step (d) is selected a target image signal in this first, second and third image signal, and this target image signal is carried out a filter, calculates one second chrominance signal; Then, execution in step (e) according to this second chrominance signal, and via this frequency demodulation program, calculates one second colourity angular frequency; Then, execution in step (f) according to this second colourity angular frequency, draws one second chromatic value.
At last, execution in step (h), with step (c) calculate and this first chromatic value and step (f) calculate and this second chromatic value multiply by totalling behind one first weights and one second weights respectively, represent chromatic value to calculate one.Wherein, first weights and second weights can suitably be adjusted weight proportion according to the color distribution situation of reception image.Because the present invention adopts the image treatment method of prior art and above-mentioned image treatment method simultaneously, therefore image treatment method of the present invention more can be handled the shades of colour distribution situation in raw video.
The present invention provides a kind of image-processing circuit in addition, in order to the image of processing based on SECAM system, and can solve the unusual color lump phenomenon that processing multiple-pulse pattern is produced in the prior art.
According to a specific embodiment, image-processing circuit of the present invention comprises receiving element, vertical filtration module, frequency demodulation module and chroma conversion unit.Vertically filtration module is coupled to this receiving element, and the frequency demodulation module is coupled to this vertical filtration module, and the chroma conversion unit is coupled to this frequency demodulation module.
Receiving element is in order to receive this image, and this image comprises a plurality of pixels.These pixels comprise one first pixel, one second pixel and one the 3rd pixel of arranging continuously on the same vertical line at least, and respectively corresponding one first image signal, one second image signal and one the 3rd image signal.
Vertically filtration module produces one first chrominance signal according to this first image signal, this second image signal and the 3rd image signal.
The frequency demodulation module calculates one first colourity angular frequency according to this first chrominance signal.Then, the chroma conversion unit calculates one first chromatic value according to this first colourity angular frequency.
Image-processing circuit that the present invention adopts and image treatment method see through vertical filter, remove the luminance signal of frequency of oscillation near carrier angular frequencies effectively, and can not bring luminance signal into follow-up frequency demodulation program.Therefore, the present invention can solve effectively and handle the unusual color lump phenomenon that the multiple-pulse pattern is produced in the prior art.In addition, prior art and image processing mode of the present invention have been merged in the present invention and then see through the chromatic value weighting scheme.Thereby the present invention dynamically adjusts the weights ratio, to handle shades of colour distribution situation in the image widely.
Description of drawings
Fig. 1 is the waveform sketch map of general CVBS.
Fig. 2 is the functional block diagram of the image-processing circuit in the prior art.
Fig. 3 (a) is the waveform sketch map of the CVBS of multiple-pulse pattern.
Fig. 3 (b) handles the image that is shown behind the CVBS among Fig. 3 (a) for the image-processing circuit of prior art.
Fig. 4 is the functional block diagram of the image-processing circuit of first specific embodiment according to the present invention.
Fig. 5 is the color distribution sketch map according to a specific embodiment.
Fig. 6 is the color distribution sketch map according to another specific embodiment.
Fig. 7 (a) is the functional block diagram of the image-processing circuit of second specific embodiment according to the present invention.
Fig. 7 (b) is the spectrogram according to the image signal of another specific embodiment.
Fig. 8 is the functional block diagram of the image-processing circuit of the 3rd specific embodiment according to the present invention.
[primary clustering symbol description]
1,3,5,7: image-processing circuit
16,36,58,79: the chroma conversion unit
13,30,50,70: receiving element falls into the shape filter unit at 53: the first
14,34,74: frequency demodulation module 32,52: vertical filtration module
11: filter unit 55: the chroma edge detecting module
57: luminance edges detecting module 56: the colourity weighting block
552: the first judging units of 550: the first subtrators
562: the second setup units of 560: the first setup units
570: the second subtrators of 564: the three setup units
Judging unit fell into the shape filter unit at 75: the second in 572: the second
77: the three sunken 78: the three subtrators of shape filter unit
76: 760: the four setup units of luminance weighted module
10: initial prompting signal 12: color burst
S1: first image signal 20,22,24: color lump
S2: the second image signal S3: the 3rd image signal
V1: the first chromatic value V2: second chromatic value
V3: the 3rd chromatic value V4: the 4th chromatic value
V0: represent chromatic value Y1: first brightness value
Y2: the second brightness value Y3: the 3rd brightness value
Y4: the 4th brightness value Y5: the 5th brightness value
Y0: representative luminance value 51: bandpass filtering unit
Embodiment
Be divided into five main paragraphs narrations in order to let execution mode of the present invention be easier to understand: the first, bidimensional image processing method of the present invention; The second, why the bidimensional image processing method can solve the unusual color lump problem of multiple-pulse pattern; Three, the problem that run into of bidimensional image processing method; Four, the weight setting of colourity; Five, the weight setting of brightness.
The image treatment method of prior art is handled single pixel, to calculate chromatic value.Yet image treatment method of the present invention is handled at least three pixels that are arranged in continuously on the same vertical line, to calculate chromatic value.Therefore, in following explanation, the image treatment method of prior art is referred to as the one dimension image treatment method, image treatment method of the present invention is referred to as the bidimensional image processing method.
The first, bidimensional image processing method of the present invention
See also Fig. 4, Fig. 4 is the functional block diagram of the image-processing circuit 3 of first specific embodiment according to the present invention.As shown in Figure 4, image-processing circuit 3 comprises receiving element 30, vertical filtration module 32, frequency demodulation module 34 and chroma conversion unit 36 in order to carry out bidimensional image processing method of the present invention.Vertically filtration module 32 is coupled to receiving element 30, in order to the luminance signal in the filtering image signal, to stay chrominance signal.Frequency demodulation module 34 is coupled to vertical filtration module 32, in order to chrominance signal is solved chrominance frequency.Chroma conversion unit 36 is coupled to frequency demodulation module 34, in order to convert chrominance frequency into chromatic value.
Receiving element 30 is followed the coded image of SECAM-system standard in order to reception, and this image comprises at least three pixels that are arranged in continuously on the same vertical line, and these pixels are the corresponding first image signal S1, the second image signal S2 and the 3rd image signal S3 respectively.Vertically filtration module 32 calculates chrominance signal according to the first image signal S1, the second image signal S2 and the 3rd image signal S3; Frequency demodulation module 34 calculates the colourity angular frequency according to chrominance signal, and then the first chromatic value V1 of corresponding colourity angular frequency is found out in chroma conversion unit 36.
Vertically filtration module 32 comprises first multiplication unit 320 and adder unit 322.First multiplication unit 320 multiply by first multiple, second multiple and triple respectively with the first image signal S1, the second image signal S2 and the 3rd image signal S3, and wherein the summation of first multiple, second multiple and triple is zero.
The signal of Sequential Color and Memory system has following characteristic: three pixels of vertically arranging continuously in its pairing three image signals, must have two phase difference of pi of phase place and other (supposing that three pixels have identical brightness and colourity) of one.In order to let above-mentioned explanation be easier to understand, lift the handling process of the vertical filtering module 32 of an example explanation.The corresponding respectively first image signal S1, the second image signal S2 and the 3rd image signal S3 of three pixels of vertically arranging continuously is:
S1＝Y+Sin[(ωc+Δω)*t+φ1]，
S2＝Y+Sin[(ωc+Δω)*t+φ1]，
S3＝Y+Sin[(ωc+Δω)*t+φ1+π]。
N1*S1+N2*S2+N3*S3＝Y*(N1+N2+N3)+N1*Sin[(ωc+Δω)*t+φ1]+N2*Sin[(ωc+Δω)*t+φ1]+N3*Sin[(ωc+Δω)t+φ1+π]＝Y(N1+N2+N3)+(N1+N2-N3)Sin[(ωc+Δω)*t+φ1]
Wherein, first Y* (N1+N2+N3)=0 of following formula result represents luminance signal by filtering, yet on behalf of chrominance signal, another (N1+N2-N3) * Sin [(ω c+ Δ ω) * t+ φ 1] ≠ 0 be retained.Therefore, vertically filtration module 32 can leach the chrominance signal of image signal effectively, is first chrominance signal to call this chrominance signal in the following text.
Then, first chrominance signal that vertical filtration module 32 is leached inputs to frequency demodulation module 34, and frequency demodulation module 34 can parse the first colourity angular frequency (Δ ω) of first chrominance signal.The operating principle of frequency demodulation module 34 is identical with frequency demodulation module 14, so repeat no more.
Be noted that here based on the logic that the phase angle should change continuously, the absolute value of Δ ω should be not more than the π radian, just be not more than 180 degree, reach (π) between the radian so the difference between the two phase place angle of corresponding two pixels should fall within the π radian.In other words, if phase difference greater than the π radian, then need deduct 2 π radians; If phase difference is less than (π) radian then need add 2 π radians.
The above-mentioned image treatment method that image-processing circuit of the present invention and enforcement thereof have been described.Yet why the one dimension image treatment method can solve unusual color lump in multiple-pulse pattern (multi-burst pattern), and bidimensional image processing method of the present invention just can not solve unusual color lump, asks for an interview following explanation.
The second, why the bidimensional image processing method can solve the unusual color lump problem of multiple-pulse pattern
For instance, in chequered with black and white multiple-pulse pattern, target image signal S ' should be:
S′＝Y+Sin[(ωc)*t]
Because of chequered with black and white multiple-pulse pattern does not have color, be zero so represent the colourity angular frequency (Δ ω) of colourity, and represent the Y of luminance signal in the multiple-pulse pattern, to change with CF.In this example, luminance signal is with frequency (the ω c+ δ ω) change in oscillation near carrier angular frequencies, and wherein δ ω represents small difference on the frequency.
In general, luminance signal is the signal of low frequency, and its frequency is much lower compared with carrier angular frequencies (ω c), and therefore the mode of filter unit 11 usefulness of convention shown in Figure 2 filtration particular frequency range just can the filtering luminance signal.Yet if the frequency of oscillation of luminance signal is near carrier frequency ω c, traditional filter unit 11 (shown in Figure 2) then can't filter luminance signal effectively.That is can be takeed in the image-processing circuit of convention with the luminance signal of specific concussion frequency change is chrominance signal, in chequered with black and white multiple-pulse pattern, has produced unusual color lump and cause, shown in Fig. 3 (b).
As far as the multiple-pulse pattern, vertical filtration module 32 of image-processing circuit 3 is not to remove luminance signal through the mode of frequency filtration, but utilizes the characteristic of Sequential Color and Memory system that luminance signal is removed.Therefore, image-processing circuit of the present invention can be handled the luminance signal of change in oscillation effectively, can't in chequered with black and white multiple-pulse pattern, decode unusual color lump.
Three, the problem that run into of bidimensional image processing method
In following two kinds of situation: A, stride look district situation and B, the uneven situation of color distribution, the bidimensional image processing method that the image-processing circuit 3 among Fig. 4 is carried out has more unfavorable result and produces, with being described below respectively.
A, stride the situation (as shown in Figure 5) in look district
The prerequisite of above-mentioned bidimensional image processing method is that three pixels of vertically arranging continuously all are with same frequency oscillation.In fact, under the situation of striding the look district, that is to say that as if look area edge position the frequency of oscillation of one of them pixel of three pixels can be different with other frequency of oscillation of two pixels between three pixels that adopted.
For instance, second pixel in three pixels of arranging continuously and the 3rd pixel be as if the look district that lays respectively at different colours, and then the pairing image signal of first, second and third pixel (S1 ', S2 ', S3 ') is respectively:
S1′＝Y+Sin[(ωc+Δω1)*t+φ1]，
S2′＝Y+Sin[(ωc+Δω1)*t+φ1]，
S3 '=Y+Sin [(ω c+ Δ ω 2) * t+ φ 1+ π], wherein Δ ω 2 ≠ Δ ω 1.
After the calculating through the vertical filtration module 32 among Fig. 4, can obtain the item of Sin [(ω c+ Δ ω 1) * t]+Sin [(ω c+ Δ ω 2) * t], be equivalent to have frequency of oscillation and be (ω c+ Δ ω 1) signal with (ω c+ Δ ω 2).And then frequency demodulation module among Fig. 4 34 and the chromatic value that chroma conversion unit 36 calculates will be Δ ω 1 and the mixing of Δ ω 2 pairing colors, but not Δ ω 1 or Δ ω 2 pairing colors.That is to say, under the situation of striding the look district, can solve the color that differs from adjacent dichromatism district, and form color lump on the look area edge, as shown in Figure 5.The dichromatism district that Fig. 5 is neighbouring and color is different, and the edge between the dichromatism district has formed unusual color lump 22.
B, color distribution inhomogeneous (as shown in Figure 6)
In above-mentioned bidimensional image processing method, in pairing three image signals of arranging continuously of three pixels, phase place and other two phase difference of pi (characteristic of SECAM-system) of one must be arranged.But, if the front end color distribution of horizontal scanning linear irregular (like the color distribution situation of the left side of Fig. 6) will cause extra phase difference.For instance, first, second and third horizontal scanning linear respectively the corresponding image signal (S1 ", S2 ", S3 ") are:
S1″＝Y+Sin[(ωc+Δω)*t+φ1]，
S2″＝Y+Sin[(ωc+Δω)*t+φ2]，
S3″＝Y+Sin[(ωc+Δω)*t+φ3+π]，
Wherein+π is the standard phase difference of SECAM-system, and φ 1, φ 2, φ 3 are because the inhomogeneous phase difference that causes of front end color distribution.
Therefore, causing the phase difference of these continuous three pairing image signals of pixel is not simple difference π.If just the phase difference of φ 1, φ 2, φ 3+ π is 0; Behind vertical filtration module 32 among input Fig. 4; Result of calculation will be 0; Can be considered the signal of Sin [(ω c+ (ω c)) * t], just the absolute value of colourity angular frequency equals carrier angular frequencies, but in fact the colourity angular frequency is not equal to carrier angular frequencies.According to the colourity angular frequency of SECAM and the table of comparisons of chromatic value (color), Δ ω=-corresponding color of ω c is about peony.In a word, so may let the bidimensional image processing method calculate unexpected color, as shown in Figure 6.Fig. 6 is the look district that the left and right sides is adjacent and color is different, and the color distribution in the left side look district is inhomogeneous, and unusual color lump 24 has appearred in the edge that causes left right-hand part look district to join.
Four, the weight setting of colourity
In order to improve the above-mentioned image processing problem that is produced in look district and the uneven situation of color distribution of striding, the present invention sees through the setting of weights and deals with.In more detail; See also Fig. 7 (a); Same image obtains first chromatic value via bidimensional image processing mode of the present invention (that is via vertical filtration module 52, frequency demodulation module 54 and chroma conversion unit 58), and obtains second chromatic value via one dimension image treatment method (that is via bandpass filtering unit 51, frequency demodulation module 54 and chroma conversion unit 58).Then, then first chromatic value and second chromatic value multiply by totalling behind first weights and second weights respectively, to calculate the representative chromatic value.Utilize the adjustment of first weights and second weights, set the importance of first chromatic value and second chromatic value.Thus, just present situation, increased the use elasticity of image treatment method of the present invention applicable to different colours.
See also Fig. 7 (a), image-processing circuit 5 comprises colourity weighting block 56.Colourity weighting block 56 is coupled to chroma conversion unit 58, and the first chromatic value V1 and the second chromatic value V2 multiply by totalling behind first weights and second weights respectively, represents chromatic value V0 to calculate.
1, first setup unit
Uneven explanation can be known according to above-mentioned thirdly middle color distribution; Angular frequency rate variance between the first colourity angular frequency that the bidimensional image processing method is calculated and the carrier angular frequencies of SECAM-system may be very big, thereby first setup unit 560 can moderately be turned down first weights and heighten second weights to this kind situation.Just when the angular frequency rate variance between the carrier angular frequencies of the first colourity angular frequency and SECAM-system was very big, bidimensional image processing method of the present invention produced more unfavorable effect possibly, therefore cooperated the one dimension image treatment method to obtain preferable effect.
According to above explanation, first setup unit 560 can judge whether the first colourity angular frequency surpasses a predetermined threshold value.If judged result is for being then to set first weights less than second weights; If judged result is then set first weights greater than second weights for not.
2, second setup unit
The image that receiving element 50 is received more comprises the 4th pixel and the 5th pixel; And these pixels are arranged in vertical direction in regular turn continuously: from top to bottom be the 4th, first, second, third and the 5th pixel, and respectively corresponding the 4th, first, second, third and the 5th image signal.Vertically filtration module 52, frequency demodulation module 54 and chroma conversion unit 58 are according to the 4th image signal, first image signal and second image signal; Calculate the 3rd chromatic value V3; And, calculate the 4th chromatic value V4 according to second image signal, the 3rd image signal and the 5th image signal.
Image-processing circuit 5 of the present invention comprises chroma edge detecting module 55, is coupled to chroma conversion unit 58.Chroma edge detecting module 55 calculates chroma edge intensity (colourity difference) according to the 3rd chromatic value V3, the first chromatic value V1 and the 4th chromatic value V4 of 58 outputs of chroma conversion unit.
In logic, if chroma edge appears between first, second, third pixel, then the first chromatic value V1, the 3rd chromatic value V3 and the 4th chromatic value V4 two chromatic values wherein must have suitable difference.So the present invention judges whether to be the look area edge by chroma edge intensity (colourity difference), and the calculating of chroma edge intensity is described below.
Chroma edge detecting module 55 comprises first subtrator 550 and first judging unit 552.First judging unit 552 is coupled to first subtrator 550.After the 3rd chromatic value V3 that first subtrator 550 is exported chroma conversion unit 58 deducts the first chromatic value V1, calculate the first colourity difference, and the 4th chromatic value V4 is deducted the first chromatic value V1, calculate the second colourity difference.Whether the absolute value that first judging unit 552 is judged this first colourity difference is greater than the absolute value of this second colourity difference.If judged result is for being, then first judging unit, 552 these chroma edge intensity of setting are the absolute value of this first colourity difference; If judged result is not for, first judging unit 552 absolute value that to set these chroma edge intensity be this second colour difference then.
If two pixels of vertical adjacent arrangement belong to the look district of different colours, then corresponding respectively chromatic value has difference to a certain degree.Therefore, second setup unit 562 can be told the situation of crossing over the look district that whether is at present according to chroma edge intensity.By this, second setup unit 562 just can be set these first weights and this second weights according to the size of chroma edge intensity (colourity difference).For instance, if chroma edge intensity is the absolute value of the first colourity difference and greater than a predetermined threshold value, then very likely have the color lump edge between first pixel to the, three pixels.Therefore, the first chromatic value V1 with the bidimensional image processing method is calculated gives lower weight proportion.
3, the 3rd setup unit
Except differentiating chroma edge, the present invention also includes luminance edges in consideration.
Shown in Fig. 7 (a), image-processing circuit 5 of the present invention comprises first and falls into shape filter unit 53 and luminance edges detecting module 57.First falls into shape filter unit 53 is coupled to receiving element 50, and luminance edges detecting module 57 is coupled to first and falls into shape filter unit 53.First falls into shape filter unit 53 in order to handling first image signal, second image signal and the 3rd image signal, with the second brightness value Y2 that calculates the first brightness value Y1 of corresponding first pixel, corresponding second pixel respectively and the 3rd brightness value Y3 of corresponding the 3rd pixel.Shown in Fig. 7 (b); Curve 62 and 64 is represented the brightness composition and the chromatic component of image signal respectively; Dotted line 66 represents first to fall into shape filter unit 53 and kept most low-frequency signal (representing the brightness composition), and filtering be the signal (representative chromatic component) of a frequency range at center with carrier angular frequencies (ω c).
Luminance edges detecting module 57 falls into the first brightness value Y1, the second brightness value Y2 and the 3rd brightness value Y3 that shape filter unit 53 is exported according to first, calculates luminance edges intensity.
About the calculating of luminance edges intensity, luminance edges detecting module 57 is to accomplish through second subtrator and 570 and second judging unit 572.Second judging unit 572 is coupled to first subtrator 570.The first brightness value Y1 that second subtrator 570 is exported the first sunken shape filter unit 53 deducts the second brightness value Y2, calculates first luminance difference, and the 3rd brightness value Y3 is deducted the second brightness value Y2, calculates second luminance difference.
Whether the absolute value that second judging unit 572 is judged first luminance difference that second subtrator 570 is exported is greater than the absolute value of second luminance difference.If judged result is for being, then second judging unit, 572 setting luminance edges intensity are the absolute value of first luminance difference; If judged result is not, then second judging unit, 572 setting luminance edges intensity are the absolute value of second luminance difference.
If two pixels of vertical adjacent arrangement belong to the zone of different brightness, then corresponding respectively brightness value has difference to a certain degree.Therefore, the 3rd setup unit 564 can be told the situation of crossing over different luminance areas that whether is at present according to luminance edges intensity.For instance, if luminance edges intensity is the absolute value of first luminance difference and greater than a predetermined threshold value, then very likely have luminance edges between first pixel and second pixel.Therefore, the first chromatic value V1 with the bidimensional image processing method is calculated gives lower weight proportion.
Five, the weight setting of brightness.
Above-mentioned declarative description calculating and the setting of weights of chromatic value.For for the sunken shape filter that leaches luminance signal, also can follow similar weights notion, set the scope of filtering frequency.
In order to obtain preferable brightness performance, the present invention provides another kind of image-processing circuit 7, sees also Fig. 8.As shown in Figure 8, image-processing circuit 7 comprises second and falls into shape filter unit the 75, the 3rd sunken shape filter unit 77 and luminance weighted module 76.The second sunken shape filter unit 75 and the 3rd falls into shape filter unit 77 and all is coupled to receiving element 70, and luminance weighted module 76 is coupled to second and falls into shape filter unit 75 and the 3rd sunken shape filter unit 77.
Receiving element 70 is in order to receive the image based on the Sequential Color and Memory system standard; This image comprises a plurality of pixels; These pixels comprise first pixel, second pixel and the 3rd pixel of arranging continuously on the vertical direction at least, and difference corresponding first image signal, second image signal and the 3rd image signal.
Second fall into shape filter unit 75 and the 3rd fall into shape filter unit 77 with the wherein image signal in first, second and third image signal as a target image signal; And handle this target image signal with first frequency scope and second frequency scope respectively; To calculate the 4th brightness value Y4 and the 5th brightness value Y5 respectively, wherein the second frequency scope contains and greater than the first frequency scope.
Luminance weighted module 76 multiply by totalling behind the 3rd weights and the 4th weights respectively with the 4th brightness value Y4 and the 5th brightness value Y5, to calculate the representative luminance value Y0 of corresponding target image signal.Wherein, luminance weighted module 76 be according to one dimension image treatment method and bidimensional image processing method calculate respectively and chromatic value, set the size of the 3rd weights and the 4th weights, and detailed handling process will be described in down.
Image-processing circuit 7 comprises chroma conversion unit 79 in addition and is coupled to the 3rd subtrator 78 of chroma conversion unit 79.After the first chromatic value V1 that the 3rd subtrator 78 is exported chroma conversion unit 79 deducts the second chromatic value V2, calculate the colourity difference.
Luminance weighted module 76 comprises the 4th setup unit 760, is coupled to the 3rd subtrator 78, and the size of the absolute value of the colourity difference of being exported according to the 3rd subtrator 78 is set the 3rd weights and the 4th weights.If the absolute value of colourity difference is bigger, then adopt narrower frequency range, so set the 3rd weights greater than the 4th weights as far as possible; Otherwise,, then adopt wider frequency, so set the 3rd weights less than the 4th weights if the absolute value of colourity difference is littler as far as possible.
Comprehensively above-mentioned; Can know by above-mentioned first and second; Bidimensional image processing method of the present invention (image-processing circuit 3 of Fig. 4) sees through vertical filtering and calculates, the unusual color lump that the one dimension image treatment method (image-processing circuit 1 of Fig. 2) of solution prior art is calculated in the multiple-pulse pattern.
Thirdly reach at the 4th and can know by above-mentioned; The present invention's (image-processing circuit 5 of Fig. 7 (a)) sees through the mode of weight setting; With the one dimension image method of bidimensional image processing method of the present invention and convention calculate respectively and first chromatic value and second chromatic value, moderately give a weight proportion.Therefore, in the various colors distribution situation, the present invention all can calculate believable chromatic value.
Can know by above-mentioned the 5th; The present invention's (image-processing circuit 7 of Fig. 8) judges present color distribution situation according to the difference of first chromatic value and second chromatic value, and then in falling into the shape filter, can suitably set in order to filter out the filtration frequency range of luminance signal.Therefore, under the different colours distribution situation, sunken shape filter of the present invention is separating luminance and chrominance signal all effectively.
By the detailed description of above preferred embodiment, be that hope can be known description characteristic of the present invention and spirit more, and be not to come category of the present invention is limited with the above-mentioned preferred embodiment that is disclosed.On the contrary, its objective is that hope can contain in the category that is arranged in claim of the present invention of various changes and tool equality.Therefore, the category of claim of the present invention should be done the broadest explanation according to above-mentioned explanation, contains the arrangement of all possible change and tool equality to cause it.
Claims (16)
1. image treatment method; In order to handle a image based on a SECAM system; This image comprises one first pixel, one second pixel and one the 3rd pixel of arranging continuously on the same vertical line; And respectively corresponding one first image signal, one second image signal and one the 3rd image signal is characterized in that this image treatment method comprises the following step:
(a) according to this first image signal, this second image signal and the 3rd image signal; And via a vertical filter; This first image signal, this second image signal and the 3rd image signal multiply by one first multiple, one second multiple and a triple respectively; Wherein the summation of this first multiple, this second multiple and this triple is zero; With first image signal after the multiplying, this second image signal and the 3rd image signal addition after the multiplying after the multiplying, calculate one first chrominance signal;
(b), and, calculate one first colourity angular frequency via a frequency demodulation program according to this first chrominance signal; And
(c) produce one first chromatic value according to this first colourity angular frequency.
2. image treatment method as claimed in claim 1 is characterized in that, more comprises the following step:
(d) in this first, second and third image signal, select a target image signal, and this target image signal is carried out a filter, calculate one second chrominance signal;
(e), and, calculate one second colourity angular frequency via this frequency demodulation program according to this second chrominance signal; And
(f) draw one second chromatic value according to this second colourity angular frequency.
3. image treatment method as claimed in claim 2 is characterized in that, more comprises the following step:
(h) this first chromatic value and this second chromatic value multiply by totalling behind one first weights and one second weights respectively, represent chromatic value to calculate one.
4. image treatment method as claimed in claim 3 is characterized in that, in step (h) before, comprises the following step:
(g1) judge whether this first colourity angular frequency surpasses a threshold value;
(g2) if judged result is for being then to set these first weights less than these second weights; And
(g3) if judged result for not, is then set these first weights greater than these second weights.
5. image treatment method as claimed in claim 3; It is characterized in that; The pixel of arranging continuously on these same vertical lines more comprises one the 4th pixel and one the 5th pixel; Respectively corresponding one the 4th image signal and one the 5th image signal, the 4th pixel, this first pixel, this second pixel, the 3rd pixel and the 5th pixel are arranged in regular turn continuously, and (h) comprises the following step before in step:
(g1), and, calculate one the 3rd chromatic value via this vertical filter and this frequency demodulation program according to the 4th image signal, this first image signal and this second image signal;
(g2), and, calculate one the 4th chromatic value via this vertical filter and this frequency demodulation program according to this second image signal, the 3rd image signal and the 5th image signal;
(g3), and, calculate a colourity edge strength via a colourity edge detection program according to the 3rd chromatic value, this first chromatic value and the 4th chromatic value; And
(g4) set these first weights and this second weights according to the size of this chroma edge intensity.
6. image treatment method as claimed in claim 3 is characterized in that, (h) comprises the following step before in step:
(g1) respectively this first image signal, this second image signal and the 3rd image signal are carried out one and fall into the shape filter, to calculate to one first brightness value that should first pixel, to one second brightness value that should second pixel and to one the 3rd brightness value that should the 3rd pixel;
(g2), and, calculate a luminance edges intensity via a luminance edges detection process according to this first brightness value, this second brightness value and the 3rd brightness value; And
(g3) set these first weights and this second weights according to the size of this luminance edges intensity.
7. image treatment method as claimed in claim 2 is characterized in that, more comprises the following step:
(i1) set one fall into the shape filter a first frequency scope and a second frequency scope, this second frequency scope contains and greater than this first frequency scope;
(i2) should fall into the shape filter with this first frequency scope to this target image signal execution, calculate one the 4th brightness value;
(i3) should fall into the shape filter with this second frequency scope to this target image signal execution, calculate one the 5th brightness value; And
(i4) the 4th brightness value and the 5th brightness value multiply by totalling behind one the 3rd weights and one the 4th weights respectively, with calculate to should the target image signal one represent brightness value.
8. image treatment method as claimed in claim 7 is characterized in that, (i4) comprises the following step before in step:
After this first chromatic value deducted this second chromatic value, calculate a colourity difference; And
Size according to this colourity difference is set the 3rd weights and the 4th weights.
9. image-processing circuit in order to handle the image based on SECAM system, is characterized in that this image-processing circuit comprises:
One receiving element, in order to receive this image, this image comprises one first pixel, one second pixel and one the 3rd pixel of arranging continuously on the same vertical line, and respectively corresponding one first image signal, one second image signal and one the 3rd image signal;
One vertical filtration module; Be coupled to this receiving element; Comprise: one first multiplication unit; This first image signal, this second image signal and the 3rd image signal multiply by one first multiple, one second multiple and a triple respectively, and wherein the summation of this first multiple, second multiple and triple is zero; And
One adder unit is coupled to this first multiplication unit, with this first image signal after the multiplying, this second image signal and the 3rd image signal addition after the multiplying after the multiplying, calculates one first chrominance signal;
One frequency demodulation module is coupled to this vertical filtration module, according to this first chrominance signal, calculates one first colourity angular frequency; And
One colourity converting unit is coupled to this frequency demodulation module, in order to produce one first chromatic value according to this first colourity angular frequency.
10. image-processing circuit as claimed in claim 9 is characterized in that, more comprises:
One filter unit is coupled to this receiving element, in order to calculating one second chrominance signal according to a target image signal, this target image signal system for this first, second and third image signal one of them;
Wherein, this frequency demodulation module more is coupled to this filter unit, calculates one second colourity angular frequency according to this second chrominance signal, and this chroma conversion unit more produces one second chromatic value according to this second colourity angular frequency.
11. image-processing circuit as claimed in claim 10; It is characterized in that, more comprise a colourity weighting block, be coupled to this chroma conversion unit; This first chromatic value and this second chromatic value multiply by totalling behind one first weights and one second weights respectively, represent chromatic value to calculate one.
12. image-processing circuit as claimed in claim 11 is characterized in that, this colourity weighting block comprises one first setup unit; Be coupled to this frequency demodulation module; Judge whether this first colourity angular frequency surpasses a threshold value, if then set these first weights less than these second weights; If not, then set these first weights greater than these second weights.
13. image-processing circuit as claimed in claim 11; It is characterized in that; The pixel of arranging continuously on these same vertical lines more comprises one the 4th pixel and one the 5th pixel; Respectively corresponding one the 4th image signal and one the 5th image signal, the 4th pixel, this first pixel, this second pixel, the 3rd pixel and the 5th pixel are arranged in regular turn continuously, and this vertical filtration module, this frequency demodulation module and this chroma conversion unit are according to the 4th image signal, this first image signal and this second image signal; Calculate one the 3rd chromatic value; And according to this second image signal, the 3rd image signal and the 5th image signal, calculate one the 4th chromatic value, this image-processing circuit more comprises:
One colourity edge detection module is coupled to this chroma conversion unit, according to the 3rd chromatic value, this first chromatic value and the 4th chromatic value, calculates a colourity edge strength;
Wherein, this colourity weighting block comprises one second setup unit, is coupled to this chroma edge detecting module, sets these first weights and this second weights according to the size of this chroma edge intensity.
14. image-processing circuit as claimed in claim 11 is characterized in that, more comprises:
One first falls into the shape filter unit; Be coupled to this receiving element; Be used for handling this first image signal, this second image signal and the 3rd image signal, to calculate respectively to one first brightness value that should first pixel, to one second brightness value that should second pixel and to one the 3rd brightness value that should the 3rd pixel; And
One luminance edges detecting module is coupled to this first sunken shape filter unit, according to this first brightness value, this second brightness value and the 3rd brightness value, calculates a luminance edges intensity;
Wherein, this colourity weighting block comprises one the 3rd setup unit, is coupled to this luminance edges detecting module, sets these first weights and this second weights according to the size of this luminance edges intensity.
15. image-processing circuit as claimed in claim 10 is characterized in that, more comprises:
One second falls into the shape filter unit, is coupled to this receiving element, handles this target image signal with a first frequency scope, to calculate one the 4th brightness value;
One the 3rd falls into the shape filter unit, is coupled to this receiving element, handles this target image signal with a second frequency scope, and to calculate one the 5th brightness value, wherein this second frequency scope contains and greater than this first frequency scope; And
One luminance weighted module; Be coupled to this second sunken shape filter unit and the 3rd and fall into the shape filter unit; The 4th brightness value and the 5th brightness value multiply by totalling behind one the 3rd weights and one the 4th weights respectively, with calculate to should the target image signal one represent brightness value.
16. image-processing circuit as claimed in claim 15 is characterized in that, more comprises:
One the 3rd subtrator is coupled to this chroma conversion unit, this first chromatic value is deducted this second chromatic value after, calculate a colourity difference;
Wherein this luminance weighted module comprises one the 4th setup unit, is coupled to the 3rd subtrator, sets the 3rd weights and the 4th weights according to the size of this colourity difference.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009100090134A CN101808249B (en) | 2009-02-12 | 2009-02-12 | Image processing circuit and image processing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009100090134A CN101808249B (en) | 2009-02-12 | 2009-02-12 | Image processing circuit and image processing method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101808249A CN101808249A (en) | 2010-08-18 |
| CN101808249B true CN101808249B (en) | 2012-06-27 |
Family
ID=42609834
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2009100090134A Expired - Fee Related CN101808249B (en) | 2009-02-12 | 2009-02-12 | Image processing circuit and image processing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101808249B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0616474A2 (en) * | 1993-03-17 | 1994-09-21 | SANYO ELECTRIC Co., Ltd. | Television circuit utilizing color burst signal |
| CN1296704A (en) * | 1999-03-08 | 2001-05-23 | 松下电器产业株式会社 | Color demodulating device |
| CN1379958A (en) * | 1999-10-13 | 2002-11-13 | 汤姆森许可公司 | Digital and analog television signal digitization and processing device |
| CN1599465A (en) * | 2003-09-17 | 2005-03-23 | 松下电器产业株式会社 | Color demodulation circuit |
-
2009
- 2009-02-12 CN CN2009100090134A patent/CN101808249B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0616474A2 (en) * | 1993-03-17 | 1994-09-21 | SANYO ELECTRIC Co., Ltd. | Television circuit utilizing color burst signal |
| CN1296704A (en) * | 1999-03-08 | 2001-05-23 | 松下电器产业株式会社 | Color demodulating device |
| CN1379958A (en) * | 1999-10-13 | 2002-11-13 | 汤姆森许可公司 | Digital and analog television signal digitization and processing device |
| CN1599465A (en) * | 2003-09-17 | 2005-03-23 | 松下电器产业株式会社 | Color demodulation circuit |
Non-Patent Citations (1)
| Title |
|---|
| JP特开2005-57639A1 2005.03.03 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101808249A (en) | 2010-08-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102761766B (en) | Method for depth map generation | |
| CN1874528B (en) | Method and apparatus for improving quality of composite video signal and method and apparatus for decoding composite video signal | |
| US10321022B2 (en) | Signal processing device to generate low noise image signal in low light intensity environment | |
| CN1045209A (en) | The coding/decoding method of encoded video | |
| CN102918831A (en) | Resolution evaluation device, image processing apparatus, and image display apparatus | |
| CN1812597B (en) | Digital video signal processing apparatus and method for adaptive y/c separation | |
| CN1020325C (en) | Apparatus for combining and separating constituent components of video signal | |
| CN105451094A (en) | Subtitle color adjusting method and device | |
| JP2005027271A5 (en) | ||
| US8754989B2 (en) | Signal processing apparatus and control method thereof | |
| CN101808249B (en) | Image processing circuit and image processing method | |
| JPH06121335A (en) | Separation circuit of luminance and chrominance | |
| US5208662A (en) | Method of encoding color television signal and a circuit therefor | |
| US7355655B2 (en) | Systems and methods for separating luma and chroma information in a composite video signal | |
| US8421924B2 (en) | Image processing circuit and image processing method | |
| CN111429381B (en) | Image edge enhancement method and device, storage medium and computer equipment | |
| CA2093231A1 (en) | Side panel signal processor for a widescreen television system | |
| CN104717401B (en) | A kind of method and device for removing singular point noise | |
| CN104967761A (en) | Color gamut coupling method | |
| CN101909214B (en) | Image processing circuit and method | |
| JP2003348609A5 (en) | ||
| CN105631802B (en) | A kind of method and device of digital image format conversion | |
| CN101980543B (en) | Two-dimensional comb filter and luminance and chrominance separation method | |
| CN1179576C (en) | Color transmission system discrimination circuit in television set | |
| JP2008160419A (en) | Motion detector |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20120627 Termination date: 20190212 |