CN1248486C - Signal peaking - Google Patents
Signal peaking Download PDFInfo
- Publication number
- CN1248486C CN1248486C CN 00806349 CN00806349A CN1248486C CN 1248486 C CN1248486 C CN 1248486C CN 00806349 CN00806349 CN 00806349 CN 00806349 A CN00806349 A CN 00806349A CN 1248486 C CN1248486 C CN 1248486C
- Authority
- CN
- China
- Prior art keywords
- signal
- peaking
- vision signal
- frequency coefficient
- vision
- 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
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
- H04N19/186—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being a colour or a chrominance component
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/50—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
- H04N19/503—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving temporal prediction
- H04N19/51—Motion estimation or motion compensation
- H04N19/527—Global motion vector estimation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/85—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using pre-processing or post-processing specially adapted for video compression
- H04N19/86—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using pre-processing or post-processing specially adapted for video compression involving reduction of coding artifacts, e.g. of blockiness
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N5/00—Details of television systems
- H04N5/14—Picture signal circuitry for video frequency region
- H04N5/20—Circuitry for controlling amplitude response
- H04N5/205—Circuitry for controlling amplitude response for correcting amplitude versus frequency characteristic
- H04N5/208—Circuitry for controlling amplitude response for correcting amplitude versus frequency characteristic for compensating for attenuation of high frequency components, e.g. crispening, aperture distortion correction
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Picture Signal Circuits (AREA)
- Processing Of Color Television Signals (AREA)
Abstract
A method and apparatus for the selective peaking of portions of a video signal is provided. The video signals, such as those transmitted in compressed form, in accordance with MPEG Standards, can be presented as blocks of data encoded using Discrete Cosine Transforms (DCT). The Discrete Cosine Transform coefficients corresponding to the chrominance values of the signal can be obtained and analyzed. The analysis and corresponding amounts of enhancement (peaking) can be performed on a block-by-block, I-frame by I-frame, basis to selectively peak portions of a video signal. If the DCT coefficients for a portion of the signal exceed or fall within a pre-selected value or range, that portion of the signal can be subjected to peaking. For example, if the high frequency horizontal DCT coefficients exceed (or fall within) a pre-set threshold value or range, a horizontal peaking circuit can be set to peak that portion of the signal. The signal processor can be set for gain in proportion to how much the coefficients exceed the threshold. A similar approach can be applied in the vertical spatial direction.
Description
The background of invention
The present invention relates generally to the vision signal processing, relate more specifically to improving one's methods of peaking (peaking) signal (particularly encoding digital video signal).
Usually before showing outputting video signal, this signal is carried out peaking (signal enhancing), thereby increase the acutance of display image to strengthen high frequency edge.Conventional vision signal generally is made of brightness and color component.Be often expressed as Y-signal or signal component during luminance signal (or component of signal), and be often expressed as P signal or signal component during carrier chrominance signal (or component of signal).Vision signal generally comprises vertically, level, intersection and the details of time (detail).Though might use peaking in any one direction (for example, horizontal or vertical direction), such signal strengthens general to be carried out in a horizontal direction.
Conventional peaking system adopts the HFS of selecting signal such as the filter of high pass filter usually.General TV acutance control is carried out horizontal filtering to luminance signal.Output with high pass filter is added on the source video signal then.If this source video signal comprises high fdrequency component, then these components will be exported from high pass filter and can carry out amplitude adjusted according to required peaking amount, and return when being added on this source video signal at these components, the amplitude of this high fdrequency component of this source video signal will be promoted (increase).This clean effect is to come " peaking " or " sharpening " this image by the slope that increases corresponding to the signal at the edge of display image.Therefore, this display image looks after honeybeeization more clear or clearer.Also can use contrary the processing and reduce high fdrequency component and " soft " this image in fact.
Picture signal is carried out digital coding usually so that transmission is also decoded subsequently so that show.For example, motion picture expert group is the ISO tissue that is provided for the standard (mpeg standard) of digital form compression and store video, audio frequency and animation.MPEG-1 is standard audio and the video code model that is used for such as the low-density storage medium of CD-ROM CD-Video or the like.MPEG-2 is the standard coding format that is used for broadcast video.In MPEG, this compression method comprises I frame or frame interior.A frame is meant a complete TV image, and it can be formed by a plurality of, such as the capable field of odd-numbered and even-numbered capable.The I frame be in the mpeg standard with the frame of video of the irrelevant in-line coding of other frame of video.Repeat the I frame at interval to refresh this coded sequence with rule.Between any two I frames is P frame (predictive frame) and B frame (bi-directional predicted frames).P and B frame only comprise information or the variation between the I frame.I frame and P thereof and B frame are called GOP (image sets).
The MPEG compression method is the lossy compression method method.Encode, decode at color image and when carrying out digital reprocessing, this demonstration can be derived the feature that can not transmit.For example, when the part of image had saturated color, that part of this image may experience the remarkable enhancing of noise when adopting conventional this image of filter peaking.This blue color part for the most responsive image of eyes is especially true.Thereby when sharpen edges, conventional honeybeeization system and/or filter may increase noise beastly to the zone that is tending towards lacking color detail.
Vision signal generally comprises the noise component(s) of a certain type.This noise may appear in the practical video signal.Treatment circuit such as encoder, decoder, transmitter, analog digital converter (ADC), digital to analog converter (DAC) or the like also can be introduced noise.Because the frequency of noise is often by high pass filter, so utilize conventional peaking filter may amplify the noise envelope of this signal.It is smooth (that is, not comprising details) that the part of vision signal may be scheduled to.Increase noise envelope in subjective especially interest and troublesome in such part at signal in the time of in the part of the vision signal that appears at the even monochromatic areas of predetermined demonstration.
Advised that the whole bag of tricks improves conventional video peaking system.For example, U.S. Patent No. 4466016 discloses the TV signal filtering system and has been used to separate TV brightness and carrier chrominance signal so that the system of subsequent video signal processing.U.S. Patent No. 5416532 has been described and has been used for the adaptive video peaking circuit and the system that use with the brightness/chroma separator.These lists of references are described circuit and the system that is used to analyze nine pel arrays, uses different filters and structure ground combined horizontal and vertical peaking signal.Various other peaking and noise coring schemes are disclosed in US 5124794, US 5161016, US 4597011,4635119,5012329 and WO 89/11196.The content of all these documents is introduced in this as a reference.Believe these schemes for example owing to only can be applicable to analog signal and not can be applicable to digital signal and unsatisfactory, or can be applicable to according to the signal of MPEG or ATSC standard or unsatisfactory in the not desired location of image, increase noise and not exclusively satisfactory.
Therefore, be desirable to provide and be used for the peaking vision signal and overcome improving one's methods and system of prior art defective.
Summary of the invention
In general, according to the present invention, be provided for selecting the method and apparatus of peaking video signal portions.Can be expressed as such as the vision signal that sends with compressed format according to mpeg standard and to utilize discrete cosine transform (DCT) coded data piece.Can easily obtain also this discrete cosine transform coefficient of analytic signal.If the analyzing and testing of DCT coefficient is to suitable energy level, such as other details in the part of those features of edge of image part or image, that part that then can this signal of peaking.Signal section corresponding to smooth image can carry out peaking minimumly or not carry out peaking fully.If any, the analysis of this signal and the enhancing (peaking) of respective amount can be carried out, so that the part of selectively peaking vision signal on the basis of piece one by one.Thereby, if the DCT coefficient of the part of this signal exceeds or drops in advance in the value of selecting or scope, that part that then can this signal of peaking.For example, if the amplitude of high frequency horizontal frequency DCT coefficient surpasses (or dropping on) preset gate limit value or scope, horizontal peaking circuit that part with this signal of peaking can be set then.Can be in vertical direction or level and vertical both direction in carry out peaking.The signal processor of controlling peaking like this can be provided for exceeding with the DCT coefficient scope directly proportional gain of this thresholding.
The processor that analog signal can walk abreast and send to analog digital converter and send to the DCT coefficient that obtains this figure signal subsequently.Can analyze these coefficients and be used to control self adaptation peaking filter factor corresponding to the space piece of the given DCT piece of the appropriate section of analog signal.Also this analog signal digital signal can be transformed to, carried out peaking according to the present invention and analog signal can be back be transformed to.
In a preferred embodiment of the invention, can selectively control the peaking of colourity, brightness or these two parts of this signal according to the frequency spectrum histogram that appears at colourity energy in this picture signal and/or colour difference signal.DCT coefficient by the analysis of MPEG code signal can be determined such colourity energy.
Therefore, be desirable to provide that be used for the peaking vision signal a kind of improves System and method for and the peaking signal of the improvement that obtains from the demonstration of such system and such signal.
Another object of the present invention provides the peaking signal of the improvement that obtains from the demonstration of such system and such signal.
Another object of the present invention provides a kind of system and method that improves that is used for the peaking compressed digital video signals.
An also purpose of the present invention provides a kind of signal processor that improves that is used for the peaking vision signal.
The present invention therefore comprise several steps and these steps each other relation and be used for result and the signal finishing these steps or obtain from these steps, all these be discussed in more detail below with accompanying drawing in be illustrated, and category of the present invention will be represented in the claims.
The accompanying drawing summary
In order to understand the present invention more comprehensively, get and consult following description in conjunction with the accompanying drawings, wherein:
Figure 1A is the block diagram that comprises the video receiver of self adaptation peaking system according to a preferred embodiment of the invention;
Figure 1B is the block diagram of video receiver with Figure 1A of the bells and whistles that is used for Analog signals;
Fig. 1 C is a flow chart of discerning the step of peaking method according to a preferred embodiment of the invention;
Fig. 2 is the figure that expression is arranged in the DCT coefficient in 8 * 8;
Fig. 3 is the figure that expression is used for 8 * 8 representative DCT coefficient of smooth vision signal;
Fig. 4 is the figure of 8 * 8 DCT coefficient of the expression smooth vision signal that is used to have noise;
Fig. 5 is the figure of 8 * 8 DCT coefficient that expression has the vision signal of details;
Fig. 6 A is the demonstration graph of the screen of display video signal, and wherein this signal does not comprise noise and do not carry out peaking;
Fig. 6 B is expression corresponding to the brightness of the vision signal of the image of Fig. 6 A figure from black to leucismusization in time;
Fig. 6 C is that the amplitude of signal of presentation graphs 6B is with the figure of frequency change;
Fig. 7 A is the demonstration graph that shows after conventional peaking corresponding to the screen of the vision signal of 6B signal;
Fig. 7 B is the brightness figure from black to leucismusization in time of the display of expression Fig. 7 A of comprising conventional peaking;
Fig. 7 C is the signal frequency-domain chart corresponding to Fig. 7 B;
Fig. 8 A is the brightness figure from black to leucismusization in time that represents the noise vision signal of not peaking;
Fig. 8 B represents to be used for the signal frequency-domain of Fig. 8 A;
Fig. 9 A represents through the brightness of the noise vision signal of peaking in time from black to white variation;
Fig. 9 B represents to be used for the signal frequency-domain of Fig. 9 A;
The brightness that Figure 10 A represents the vision signal that do not have noise and do not carry out peaking in time from black to white variation;
Detected DCT coefficient in area B only among Figure 10 B presentation graphs 10A with signal of high-level frequency;
Figure 11 A represent corresponding to the signal shown in Figure 10 A but have noise vision signal brightness over time;
Among Figure 11 B presentation graphs 11A in area B detected DCT coefficient with signal of high-level frequency;
Figure 12 A represents to be used for corresponding to the amplitude of the luminance component of the vision signal of the signal shown in Figure 11 A over time;
Figure 12 B represents the DCT coefficient corresponding to the signal of the regional A of Figure 12 A;
Figure 12 C represents the DCT coefficient corresponding to the signal of the area B of Figure 12 A;
Figure 12 D represents the DCT coefficient corresponding to the signal of the zone C of Figure 12 A;
Figure 12 E is illustrated in the application peaking according to a preferred embodiment of the invention variation of the luminance component of the vision signal of Figure 12 A afterwards.
Figure 13 is the figure of the brightness DCT coefficient of the expression saturated vision signal that is used to have luminance detail and chrominance noise;
Figure 14 is the figure of colourity DCT coefficient that expression is used for the signal of Figure 13;
Figure 15 is that expression has the brightness DCT coefficient of vision signal of brightness and chrominance detail and the figure of colourity DCT coefficient;
Figure 16 A is the time dependent chart of brightness of expression vision signal;
The colourity Pb that Figure 16 B represents vision signal over time;
The colourity Pr that Figure 16 C represents vision signal over time;
The brightness that Figure 17 A represents vision signal over time;
The colourity Pb that Figure 17 B represents the noise vision signal over time;
The colourity Pr that Figure 17 C represents the noise vision signal over time;
Figure 17 D represents the signal of Figure 17 A is carried out the effect of self adaptation peaking according to a preferred embodiment of the invention;
Figure 17 E represents the signal of Figure 17 B is carried out the effect of noise filtering;
Figure 17 F represents the signal of Figure 17 C is carried out the effect of noise filtering;
The colourity that Figure 18 A represents to have the Pb of border color transition or Pr signal over time;
The colourity spectral curve of the signal of Figure 18 B presentation graphs 18A;
Figure 19 A represents the influence of noise to the signal of Figure 18 A;
How the colourity spectral curve of the signal of Figure 19 B presentation graphs 19A and such signal are subjected to the influence of noise filter;
Figure 20 A represents Pb or Pr carrier chrominance signal over time;
Figure 20 B represents that the signal to Figure 20 A carries out the effect according to self adaptation peaking of the present invention.
The detailed description of optimum implementation
Self adaptation (selection) peaking of the vision signal signal of MPEG and ATSC standard code and transmission (particularly according to) can be by obtaining or utilize this signal the DCT coefficient, make these coefficients by threshold detector or spectrum distribution/histogram and when detecting the DCT coefficient value of selection, select these parts of this signal of peaking to realize.This peaking can be even or proportional with the DCT coefficient.Thereby to be used to make peaking system peaking position be signal section on the image boundary in the analysis of DCT coefficient on the signal receiver, and not peaking is corresponding to the signal section of smooth even image.In a preferred embodiment of the invention, if the amplitude of high frequency level (or vertical) frequency DCT coefficient surpasses the preset gate limit value, level (or vertical) peaking circuit then can be set be used for exceeding the part gain proportional of this threshold value with these coefficients.Selectively, can in level and vertical direction, carry out peaking.
The digital video signal routine sends as N * N (being generally 8 * 8) piece of the continuous formsization of video pixel information.8 * 8 of each of video pixel comprise 8 row horizontal pixels and 8 row vertical pixels and can offer discrete cosine transform (DCT) encoder to derive 64 8 * 8 corresponding to the DCT coefficient.Can be on the basis of piece one by one the signal of execution analysis and respective amount strengthen so that the selection part of this signal of peaking only.
The analysis of DCT coefficient is not limited only to Threshold detection, also can carry out bandwidth detection and spectrum distribution/histogram analysis and analyze these coefficients so that carry out the border or rim detection and smooth image detection.The analysis of DCT coefficient also can be carried out a plurality of frequency bands that comprise midband.Thereby, if the DCT coefficient drops in the midband, then can carry out bandpass filtering and peaking can be based on this bandpass filtering to this vision signal.The use of low pass DCT coefficient thresholding can be used to forbid the peaking of smooth video area.Thereby the analysis of DCT coefficient can be used to utilize and can call various filtering characteristics with the proportional different amount of gain of frequency characteristic of the DCT piece of considering, leads to or low-pass characteristic such as high pass, band.
The vision signal that sends according to mpeg standard provides the DCT coefficient to be used for according to self adaptation peaking of the present invention.Yet, the signal that scheme herein also can be applied to analog video signal or not utilize the DCT value to send.In the situation of analog video signal, this analog signal also line feed is given a device, this device is transformed to this analog signal digital signal, obtains DCT coefficient and subsequent analysis at this coefficient of discussing, so that control is used for self adaptation peaking filtering corresponding to the space piece of this DCT coefficient according to the present invention.Selection part that then can this analog signal of peaking.Selectively, this signal can be changed to digital signal, can obtain the DCT coefficient, can carry out the self adaptation peaking, and this signal can utilize digital to analog converter back to be transformed to analog signal and sends to image display.
Peaking method herein also is particularly suitable for being applied to vision signal, comprises those vision signals that can be applicable to broadband high-definition television system (HDTV).Compressed before sending such as digital signal routine with unusual broadband (30MHz) signal.Comprise the generation of DCT according to the data compression of MPEG.Hold within it and to be introduced in this academic journal (1990) Rao and Yip as a reference
Discrete Cosine Transform, the principle of DCT coding is discussed among 1997 years McGraw Hill " the Digital TelevisionFundamentals " that write by Michael Robin and Michael Poulin and the WO 96/13780.
To and aspect of the present invention and embodiment be described with reference to the accompanying drawings according to following description and example.Yet each embodiment that is discussed only is used for explanation and should think to have restrictive sense.
Figure 1A is the block diagram that can be included in such as each part of the video receiver in the television set of HDTV or DVD player 100.Coded video bit stream signal 10 is from the signal source that sends this signal as compressed encoding information or utilizing known coding method to offer channel buffer 150 after to the encoded signal coding.Channel buffer 150 sends to variable-length decoder (VLD) 160 with this code signal, and this variable-length decoder (VLD) 160 extracts each different code word.VLD 160 offers DCT analyzer and peaking processor controls 180 with high level video encoded information signal 161.VLD 160 also will offer inverse quantizer 140 corresponding to the signal 162 of quantized prediction error DCT coefficient with quantized versions.VLD 160 also will offer motion compensator 130 corresponding to the signal 163 of motion vector.
Shown in Figure 1A, inverse quantizer 140 provides DCT coefficient for DCT analyzer 180, and the peaking processor controls of analyzer/processor 180 part is determined these peakedness factors according to the result that DCT analyzes.Heterogeneous peaking filter 190 is the sampling filter again that can utilize not coefficient operation on the same group during filtering operation of known type valuably, such filter is requiring a large amount of filter taps desirable particularly useful during such as the filter response that is used for interpolation to generate, and this requires less computing unit than direct conventional filter.Hold within it and be introduced in " Handbookof Digital Signal Precessing Engineering Applications " middle multiphase filter of discussing that this academic journal (1987) is as a reference edited by Doulgas F.Elliolt.
Figure 1B be receiver 100 receiver 101 that is revised as analog video signal 11 block diagram.Receiver 101 comprises analog digital (A/D) converter 151, is used to receive analog video signal 11.The A/D converter is with the digital signal 126 of conversion ' send to peaking filter 190 and send to dct processor 152, this processor from the signal 126 of conversion ' the signal 141 that obtains the DCT coefficient and will comprise these coefficients send DCT analyzer/processor 180 to, this DCT analyzer/processor 180 with signal 181 send to peaking filter 140 with the digital signal 126 of controlling when conversion ' peaking will take place.Peaking filter 190 will select peaking signal 191 to send to display 192.
In Figure 1A and 1B, the unit that comprises digital processing unit 102 can utilize discrete component to implement.For example, inverse quantizer 140 can utilize adder to implement.IDCT 130 can utilize two DCT/IDCT IC to implement.Yet current IC technology makes discrete enforcement unactual.Thereby the functional block in the big square frame of expression digital processing unit 102 can be integrated on the single IC.
But, since highly integrated, can't buy at present on the current market and can the DCT coefficient data have been exported to analyzer so that the single IC that handles by external interface.That is, be not pin on the IC-card, so be used for the internal bus of DCT coefficient because all now processing carry out on an IC.Yet this is also uncomplicated to it will be appreciated by one of skill in the art that realization.To understand: when the IC of interconnect discrete, the DCT bus can be carried out tap and the DCT coefficient data is presented to processor.
Do not use discrete component series in order to implement self adaptation peaking method, a solution just is to use the mpeg decoder able to programme of newtype, and this decoder will allow innernal CPU access DCT coefficient and determine the peaking level so that analyze.Such example is TM110 and the TM1300 medium coprocessor of PhilipsSemiconductor.
DCT analyzer and peaking processor controls can utilize dsp software to implement fully for flexibility, this also is specially adapted to the mpeg decoder able to programme that just coming into the market.By utilizing the software implementation algorithm, when being used to watch stream-type video or DVD, the application of this peaking of self adaptation can expand to home PC.
Fig. 1 C is the flow chart of self adaptation peaking method 200 according to a preferred embodiment of the invention.Or digital signal 210 is provided or at first analog signal 211 is transformed to digital signal 210 in step 212.Coded digital signal 210 and obtain the DCT coefficient in step 220 subsequently.In fact, before broadcasting or storage, carry out the coding and the generation of DCT coefficient.In the step 230 that can utilize VCR, DVD or TV to carry out, extract colourity and/or the brightness DCT coefficient or the frequency spectrum histogram of video blocks.In step 240, with these DCT coefficients and thresholding compares or this frequency spectrum histogram and distribution are compared.In step 250, according to the more selectively peaking colourity and/or the luminance signal of step 240.If this signal will show in analogue means, in step 255, can be analog signal with this signal transformation, and at this selection peaking signal of step 260 demonstration.
Describe referring to Fig. 2-20B and to be used to analyze DCT coefficient and self adaptation peaking decoded video signal according to a preferred embodiment of the invention with the method for response from the peakedness factor of peaking processor controls.
Self adaptation peaking according to the present invention adopts the DCT coefficient value to determine whether to answer the part of peaking signal and peaking to arrive what degree valuably.The DCT coefficient that is used for given video blocks is directly corresponding to frequency domain.Therefore, the peaking amount of DCT coefficient to determine in level and vertical direction, to apply of the vision signal of energy analysis and utilization DCT coding.For example, do not have edge and seldom or do not have high fdrequency component, can forbid temporarily or even peaking filtering is set comes attenuate high frequency, the amplification of the high-frequency noise that this helps to prevent that conventional peaking filter from once amplifying with respect to smooth (or piece).If enough high-frequency informations are represented in the analysis of DCT coefficient, utilize high-amplitude DCT coefficient, the self adaptation peaking circuit can be set to be used for standard or proportional signal strengthens.According to the analysis of DCT coefficient, can on the basis of piece one by one in the horizontal direction, apply this enhancing on vertical direction or the both direction.
The analysis of DCT coefficient not necessarily is limited to high-frequency detection.For example, can analyze the coefficient in the midband zone and can before showing, in this video path, introduce corresponding band pass filter, wherein adjust the coefficient (or gain) of adaptive bandpass filter according to the energy level that with intermediate range DCT coefficient is feature.Thereby, when in interested frequency band, video content not occurring, will have only the noise in the interested frequency band not to be exaggerated.
Generally, express the DCT coefficient that is arranged in 8 * 8 referring to Fig. 2-5.Each coefficient is represented a specific frequency range.As shown in Figure 2, a DCT coefficient is called the DC coefficient, and it represents the mean value of whole 8 * 8 video blocks.Each follow-up DCT coefficient is called the AC coefficient.The amplitude of AC coefficient is high more, occurs many more details in 8 * 8 video blocks.Thereby for the smooth signal that does not have noise, each AC coefficient will be 0.For example, referring to Fig. 3, represent that it is 100 DC coefficient in this case that 8 * 8 of smooth vision signal will have, and each AC coefficient will be zero.Referring to Fig. 5,8 * 8 video blocks with high details will have many high-amplitude AC coefficients.If noise occurs in vision signal, then these AC coefficients will change by noise and cause by a small margin.For example, Fig. 4 represents to be used for the DCT coefficient of smooth signal, and thereby each AC component should be zero.Thereby, owing to the noise in the signal causes each the non-vanishing AC component shown in Fig. 4.
Fig. 6 A represents television screen.Zone A is (minimum brightness) of deceiving when seeing to some X from a W level fully.Zone C is white (maximum brightness) when seeing to some Z from a Y level fully.Area B is transition between regional A and the C or edge and expands to a y from an x.Area B is shown as continuous change gray value.
The brightness that Fig. 6 B represents the raster scan line in time from zero (deceiving) to the maximum variation of (in vain).Regional A among Fig. 6 B is corresponding to the black region in the screen of Fig. 6 A, and area B is corresponding to edge or gray area, and zone C is corresponding to white portion.
Fig. 6 C represents the scan line in the frequency domain on the frequency spectrum analyser for example.Zone A and C are corresponding to smooth vision signal and do not comprise component except that DC, and promptly all AC values all are zero.Therefore, regional A and zone C are shown as a little on the initial point of Fig. 6 C.The edge transition area B comprises the spectrum component from DC to a certain cut-off frequency (fc).Thereby the AC value of this signal is corresponding to the edge between regional A and the C on the screen, and has the amplitude greater than zero.
Fig. 7 A is illustrated in the variation of the image in the screen of Fig. 6 A after the signal that the peaking that utilizes conventional high pass filter strengthens Fig. 6 B.Evidence shows: the signal transition area B diminishes.Shown in Fig. 7 B, the slope at the edge between black region A and the white portion C is steepen owing to the peaking in this zone.Referring to Fig. 7, can understand that black and white portion (those zones on the DC) remain unchanged and component that the edge from area B obtains becomes the component of peaking, shown in the peak shape in the frequency spectrum.
Fig. 8 A represents corresponding to Fig. 6 B but has the scan line of the signal of noise.Fig. 8 B represents frequency domain, wherein has: come from the component on the DC of black and white portion A and C respectively; Come from the component of (fc) from DC to cut-off frequency on edge transition area B and the noise plane that is assumed to the white noise that has smooth spectrum component on the frequency band.Yet note: this noise also can be paroxysmal in time or concentrate in some frequency band.Yet when vision signal comprised white noise, optimal viewing arrived according to advantage of the present invention.
The scan line of the signal of Fig. 9 A presentation graphs 8A after carrying out conventional peaking.The peaking result who causes owing to white noise makes the noise envelope 90 of peaking signal of Fig. 9 A greater than the noise envelope 80 of the not peaking signal of Fig. 8 A.Thereby, though the edge between black region A ' and the white portion C ' becomes precipitous, be increased in the noise of regional A ' and the middle appearance of C '.As mentioned above, the noise that occurs in smooth video area can make us unhappy especially.
The signal frequency-domain of Fig. 9 B presentation graphs 9A.Can find: for those spectrum components the peaking scope that obtains from high pass filter, the noise plane promotes.Yet, because this noise spatially appears on the whole grating scan line, so even in smooth video section, also will strengthen (peaking) this noise corresponding to regional A ' and C '.
The advantage of Figure 10-12 expression self adaptation peaking according to a preferred embodiment of the invention.Figure 10 A represents not have the signal graph of signal of Fig. 6 B of noise.Figure 10 B represents from the DCT coefficient of the scan line acquisition of Figure 10 A.Shown in Figure 10 B, the DC DCT coefficient table value of being shown a of regional A and C.Fringe region B causes the spectral content on the desired frequency band shown in the coefficient b.Thereby, during the DCT coefficient that in analyzing edge region B, spatially generates, detect and high-frequency energy (DCT coefficient) occurs.
Figure 11 B represents to be used for the signal scanning line corresponding to Figure 10 A, promptly is used to have the scan line of the noise signal of noise envelope 81.As shown in Figure 11 B, white noise makes the amplitude value of the being increased to a ' a little in the frequency domain of all DCT coefficients, and this is smooth because of white noise on this frequency band.Yet high frequency DCT coefficient will be only just occurs when the signal of handling corresponding to area B, and this is to preset thresholding because will be lower than corresponding to the amplitude of the DCT coefficient of regional A and C.Thereby when the signal of handling corresponding to regional A and C, not by peaking, and this signal will be by peaking corresponding to the part of fringe region B with these parts of having only DC coefficient and this signal.
Thereby Figure 10 A, 10B, 11A and 11B represent how to implement the notion of peaking vision signal according to a preferred embodiment of the invention, at first do not consider the situation (Figure 10 A and 10B) of signal additional noise, consider to have among Figure 11 A and the 11B situation of noise subsequently.Because shown exemplary source vision signal only comprises horizontal transition and does not comprise vertical detail, so the horizontal DCT coefficient corresponding to horizontal DC only is discussed.Yet this notion can easily expand to and comprise vertical DCT coefficient, to strengthen vertical detail and diagonal angle part.Thereby, utilize high frequency DCT coefficient only to appear in the details area (area B) opposite, and only when the signal of handling corresponding to area B, just carry out peaking with smooth video (regional A and C), only amplify the noise in the transitional region.So noise ratio lifting flat site A in the zone and the noise plane among the C are subjective more pleasant.
Figure 12 A represents the noise vision signal corresponding to the signal of Figure 11 A.Figure 12 B represents that the amplitude of the DCT coefficient of regional A (some w is to some x) is lower than thresholding amplitude th.Thereby, will drop under the DCT coefficient thresholding and this system will not carry out peaking to this part of this signal corresponding to the DCT coefficient of the noise among the regional A.Referring to Figure 12 C, the DCT coefficient of area B (some x to y) surpasses thresholding amplitude th.Thereby, will carry out peaking to this part of this vision signal.Referring to Figure 12 D, because the noise component(s) of zone C is not enough to the DCT coefficient is risen on this threshold value, so will this part of this signal not carried out peaking.
Figure 12 E represents according to a preferred embodiment of the invention the signal of Figure 12 A is carried out the self adaptation peaking.Zone A ' will be by peaking with C ', and thereby the noise envelope 82 of Figure 12 (a) will with the noise envelope 82 of Figure 12 E that is used for flat site ' identical.Area B ' in noise envelope will be bigger.Yet, the noise in this zone subjective will be more pleasant.
In a preferred embodiment of the invention, for the video signal data of a part, under the histogrammic control of frequency spectrum of the colourity energy that the selection peaking of brightness peaking filter can be in appearing at image.Can in level or vertical direction, use luminance filter, use luminance filtering in level and the vertical both direction to influence the vertical and horizontal edge of display image but be preferably in.
In a preferred embodiment of the invention, carrier chrominance signal can selectively be carried out peaking.In the situation of mpeg compressed video signal, such as by analyzing colourity I frame DCT coefficient, can determine the frequency spectrum histogram of the colourity energy in the part of present image.The I frame that is also referred to as frame interior be with mpeg standard in the frame of video of the irrelevant in-line coding of other frame of video.Repeat the I frame at interval to refresh coded sequence with rule.Frame refers to a complete television image and can be made of the interlaced field that odd-numbered is capable and even-numbered is capable.In NTSC or other interlaced scan tv images, a field refers to the row of one group of odd or even numbering, and frame comprises the row of even-numbered and odd-numbered.Between any two I frames is P frame (predictive frame) and B frame (bi-directional predicted frames).P only comprises the information that changes between relevant two I frames with the B frame.The I frame is called GOP (image sets) with its P with the B frame.
The chrominance spectrum histogram of I frame has been defined as being applied to the high reliability that the integral body of carrier chrominance signal quantizes and has measured.For example, if the chroma histogram of low frequency component mates the pattern of setting up in advance, then can estimate the degree of chromaticity quantization.Therefore, according to frequency spectrum histogram matching result, can set up strategy and apply suitable peaking amount.And, according to the comparison of frequency spectrum histogram and predetermined threshold, can refinement peaking control.
For example, if the colourity energy of histogrammic some part of chrominance spectrum in the I frame surpasses threshold value, then can in luminance path, chrominance path or this two paths, add peaking filter.In order to carry out this relatively, check the DCT coefficient of the colour difference signal of each I frame.High DCT coefficient will represent that the video signal portions (block of video data) that enough high frequency color difference energy and expression are analyzed is positioned at and should carries out on the chroma edge of peaking.On the other hand, low level DCT coefficient represents almost not have high frequency colourity energy, and can appropriately relax or eliminate fully the brightness of this " smooth " part that is used for this vision signal and/or the peaking of chrominance path.The effect of this processing is only to strengthen (peaking) to carrying out signal corresponding to this video signal portions of the details of saturated image.On the other hand, for the smooth color field or the colorful smooth field part that wherein do not need to carry out peaking, to not carry out or appropriateness is carried out peaking, this is because the peaking in that zone can increase noise beastly, especially in the zone of the red smooth field of color.
Can be valuably carry out self adaptation peaking based on the colourity energy with the I frame rate.Therefore, comparison and selection peaking will be carried out on whole GOP, and not influence this signal with higher rate.This is useful, and this is because the rate of change of filter characteristic should be low relatively speed, so that do not cause the artifact of the filter characteristic that can occur when filter characteristic changes too soon.
Thereby, utilize after decoder is reconstructed at digital luminance signal, valuably self adaptation level and vertical filter are inserted in the luminance path.These self adaptation of filter utilize frequency spectrum histogram and the amount of energy in the colour difference signal that appears in the I frame (mpeg coded video) to control.Determine the energy in frequency spectrum histogram and the colour difference signal by checking I frame DCT coefficient.Mate some model histogram or, select peaking in high frequency level or Vertical factor subsequently above in the situation of given thresholding.
According to a preferred embodiment of the invention, the self adaptation peaking based on the chromatic component of vision signal is illustrated among Figure 13-20B.Figure 13 represents to have the DCT coefficient of 8 * 8 block of pixels of the saturated vision signal of luminance detail and chrominance noise.Figure 14 represents the feature colourity DCT coefficient of 8 * 8 pixels.Figure 15 represents corresponding to the brightness DCT piece 51 of the vision signal of the part of the image 50 with brightness and chrominance detail and the DCT coefficient of colourity DCT piece 52.
The time dependent amplitude of brightness and carrier chrominance signal is shown in Figure 16 A, 16B and 16C.In the RGB territory, do not carry out Digital Signal Processing.Therefore, be the YPbPr territory at first with this signal transformation, wherein Y represents luminance signal, and Pb represents blue carrier chrominance signal, and Pr represents red degree signal.Carrier chrominance signal is two kinds of bipolar signals between the opposite color, and wherein zero is 1/2 level between two opposite colors.As in Figure 16 A-16C see that regional A and C represent not have the flat site of details, and area B is represented the edge transition zone between regional A and the C.Therefore, Figure 16 A-16C represents to have the YPbPr chart of the color vision signal of fringe region.
Figure 17 A, 17B and 17C represent not have precipitous transition and thereby do not have fringe region but have the YPbPr territory that utilizes noise envelope 710 and the color vision signal of 711 noises of representing.The time scale of Figure 17 A-C is more much bigger than the time scale of Figure 16 A-C.Figure 17 D represents the result of the signal of peaking Figure 17 A.Can understand: the slope of area B has been increased to sharp-pointed fringe region B '.But peaking can increase the noise envelope of Pb and Pr signal and thereby preferably include the chrominance noise that noise filter reduces Pb and Pr signal.
Figure 18 A represents to have the carrier chrominance signal Pb or the Pr of smooth color area A and C and edge transition area B.The frequency component of the signal of Figure 18 A is illustrated among Figure 18 B.Zone A and C are smooth, and its chrominance spectrum curve utilization point 810 is represented.The chrominance spectrum curve representation of fringe region B is a curve 811.Figure 19 A represents to have the Pb or the Pr curve of the carrier chrominance signal (that is the signal that, has noise) of flat site A, fringe region B and smooth zone C.The chrominance spectrum chart of the signal of Figure 19 B presentation graphs 19A.Point 910 expressions do not have noise or the chart of regional A and C after the noise filter effect.The chrominance spectrum chart of noise time domain A and C is considered in zone 911 expressions.Figure 19 B represents the noise envelope 914 of the noise envelope of fringe region B greater than the effect of expression noise filter.Yet the noise in the noise ratio flat site in the transitional region is troublesome more not.
Figure 20 A representation class is similar to the Pb or the Pr colourity chart of the signal with flat site A and C and transitional region B of Figure 18 A.Figure 20 B represents how to make fringe region B sharpen into B ' according to the DCT coefficient of the high fdrequency component of presentation graphs 18B.As previously discussed, because only in area B, carry out peaking, so do not increase can become noise among more bothersome regional A and the C of noise wherein.
Table 1
| Signal type | Signal level | Handle | ||||||
| Y | Pb | Pr | Peaking Y | Peaking colourity | Extend | CTI | Noise reduces | |
| The edge | The edge | The edge | High or low | Connect | Connect | Disconnect | Connect | Disconnect |
| The edge | Noise | Noise | High or low | Connect | Disconnect | Disconnect | Disconnect | Connect |
| The edge | Edge with noise | The edge | High | Connect | Connect | Connect, reduce contrast | Redness is connected | Connect |
| The edge | Edge with noise | The edge | Low | Connect | Connect | Disconnect | Redness is connected | Disconnect |
| The edge | The edge | Edge with noise | High | Connect | Connect | Connect, increase contrast | Blue look is connected | Connect |
| The edge | The edge | Edge with noise | Low | Connect | Connect | Connect, increase contrast | Blue look is connected | Disconnect |
The signal processing technology that table 1 general introduction can be adopted according to the present invention.Go out peaking brightness (Y) signal and carrier chrominance signal (P) in the edge region as shown.Also reduce noise valuably for the high level carrier chrominance signal that noise wherein occurs.It is useful extending when detecting the edge and noise occurring, but contrast may need to adjust.When detecting the edge but noise not occurring, need the color transmission to improve (CTI).In another embodiment of the present invention, when on Pb or Pr signal, detecting the edge, a peaking Y-signal.In an also embodiment of the present invention,, then select peaking will be limited to Y-signal if in Pb or Pr signal, detect noise.
As understanding from the front, self adaptation peaking method provides benefit for the demonstration of the vision signal that sends according to an embodiment of the invention.Do not need YUV to analyze carrier chrominance signal to the rgb color matrix.The DCT analyzer of Fig. 1 and peaking processor controls 180 can be the forms of signal processor, can comprise one or more inputs, signal processing unit that is used for received signal and the output that is used for output signal.This signal processing unit can be configured to hardware, firmware and maybe can utilize software to control.The signal processor that is used for carrying out according to the present invention the self adaptation peaking can append to video processing chain before display video signal.This vision signal can be analog or digital signal (MPEG or non-MPEG); Yet, when this vision signal is mpeg format, reducing the additional desired peripheral hardware of this improvement method, this is to obtain easily because of the DCT coefficient.
Television set constructed according to the invention comprises source signal processing unit, self adaptation peaking unit, peaking system and display.If this television set structure is for receiving the MPEG compressed digital video signals, then transmission unit sends to self adaptation peaking unit with DCT information and this signal is presented to the peaking system.Which part of these self adaptation peaking unit controls peaking systems and definite this signal will be carried out peaking.Subsequently, show this signal.If this television set is not configured to receive such signal, then the source signal processing unit can comprise A/D and/or D/A converter and the unit that is used to obtain the employed DCT coefficient in self adaptation peaking unit.
Processor that implement to analyze the algorithm of DCT coefficient can be programmed for makes " peaking decision " or this determines (based on the thresholding or the spectrum distribution/histogram analysis of a certain type) to carry out " filtering " or " on average " variation with " smoothly falling " filter and coefficient on the piece at a certain predetermined quantity on the basis of piece one by one.Thereby, when carrying out peaking, decision can consider that image is in the space or temporal " history ".By on signal receiver, carrying out the peaking of selecting, can send the signal that is transmitted according to the ATSC standard.
Thereby will understand: can realize the above-mentioned target known from the foregoing description effectively, and do not deviate from spirit of the present invention and category because can carry out certain some change in said method and structure, all incidents shown in description or the accompanying drawing should be interpreted as schematically and should not be construed as restrictive sense above planning to be included in.
Also should understand: following claims are planned to cover all general and particular characteristics of the present invention as herein described and can think all statements that drop on category of the present invention therebetween on language.
Claims (34)
1. be used for a kind of method of peaking vision signal, comprise:
Reception is as the streamed source video signal of packed data, and described packed data stream comprises the frequency coefficient that is used for decompressed video signal;
From then on obtain in the source video signal to be used for this vision signal the selection part frequency coefficient and be identified for the described amplitude of selecting the frequency coefficient of part;
The amplitude of frequency coefficient and predetermined threshold level or spectrum distribution are compared; With
When the amplitude of the frequency coefficient of the selection of this vision signal part equals or exceeds predetermined threshold level or mate this spectrum distribution, the selection part of this vision signal of peaking.
2. method as claimed in claim 1, wherein the step of peaking adopts the high pass filter that is set to preset frequency, and the increase in the selection of this vision signal part partly has the frequency that is higher than this preset frequency.
3. method as claimed in claim 1, wherein the step of peaking comprises and makes this vision signal through having the band pass filter of predetermined frequency band, and the increase in the selection of this vision signal part partly falls in the predetermined frequency band.
4. method as claimed in claim 1, wherein the amplitude of the degree of peaking and frequency coefficient is proportional above the scope of this predetermined threshold level.
5. method as claimed in claim 1, this predetermined threshold level wherein is set, with the selection part of determining this vision signal whether corresponding to the selection level that shows details, thereby to the selection detail section of vision signal and the relatively flat of this signal is not partly carried out peaking.
6. a signal processor is used for the peaking vision signal, and this signal processor comprises:
Be used to receive as the importation of the streamed source video signal of packed data, described packed data stream comprises the frequency coefficient that is used for decompressed video signal;
Be used for obtaining to be used for the selection frequency coefficient partly of this vision signal and the device of the amplitude of the frequency coefficient that is identified for described selection part from source video signal;
Be used for device that the amplitude of frequency coefficient and predetermined threshold level or spectrum distribution are compared; And
The device that is used for the selection part of peaking vision signal when the amplitude of the frequency coefficient of the selection of this vision signal part equals or exceeds predetermined threshold level or coupling spectrum distribution.
7. television set comprises:
Shell is equipped with signal input part, source signal processor, peaking signal processor, peaking filtering system and video display in the above;
The source signal processor, be coupled to this signal input part and be coupled to the peaking signal processor and the peaking system, this source signal processor is constructed to send to this peaking system from signal input part received signal and the frequency coefficient that sends corresponding to vision signal to the peaking signal processor and with vision signal;
The peaking signal processor, be coupled to the source signal processor on the input and on output, be coupled to the peaking system, this peaking signal processor is constructed to: from source signal processor received signal and the frequency coefficient that obtains corresponding to the selection part of vision signal, the amplitude of frequency coefficient and predetermined threshold level or spectrum distribution are compared, and when amplitude equals or exceeds this thresholding or coupling spectrum distribution, the output control signal is given the peaking system, so that this peaking system carries out the peaking operation;
The peaking system, be constructed to receive control signal from source signal processor receiving video signals with from the peaking signal processor, and come the peaking vision signal according to being included in, and will select the vision signal of peaking to send to display from the instruction in the control signal of peaking signal processor;
Display is constructed to receive the vision signal of selecting peaking and show this vision signal from the peaking system.
8. television set as claimed in claim 7, wherein the source signal processor is constructed to digital video signal and decoding and this signal of decompression of received code compression.
9. television set as claimed in claim 7, wherein peaking system comprises the heterogeneous mode filter of sampling again that is constructed to receive a plurality of filter taps.
10. television set as claimed in claim 7, wherein the source signal processor comprises analog-to-digital converter and is constructed to derive the signal frequency-domain coefficient that utilizes described converter conversion, and described frequency coefficient is offered the peaking signal processor.
11. be used for a kind of method of peaking vision signal, comprise:
Reception sources vision signal, this source video signal comprise brightness and chrominance section and comprise that as the streamed colour difference signal subsection of packed data, described packed data stream comprises the frequency coefficient that is used for decompressed video signal;
From source video signal, obtain the spectrum distribution or the colourity frequency coefficient of the selection colourity energy partly of this vision signal;
This spectrum distribution and predetermined spectrum distributed compare or the amplitude and the predetermined threshold level of frequency coefficient compared; With
The comparison that distributes according to spectrum distribution and predetermined spectrum, or when the amplitude of the frequency coefficient of the selection part of this vision signal equals or exceeds predetermined threshold level, the brightness and/or the chrominance section of the selection part of this vision signal of peaking.
12. as the method for claim 11, wherein when the colourity frequency coefficient of this colour difference signal equals or exceeds thresholding, the selection part of this vision signal of peaking.
13. as the method for claim 11, wherein obtain the spectrum distribution of the colourity energy of each I frame, and selectively carry out peaking according to this spectrum distribution.
14. as the method for claim 11, wherein colourity I frame frequency domain coefficient and thresholding compare, and relatively carry out peaking according to this.
15. method as claim 11, wherein this source video signal comprises the signal subsection of expression I frame, and when spectrum distribution or frequency coefficient represent that the colourity energy of the predetermined portions that the chrominance spectrum in the I frame distributes exceeds threshold value, in luminance path, implement peaking.
16. as the method for claim 11, wherein the signal subsection that expression is had a border between the part of display of aberration is implemented peaking, and less or do not implement peaking to the signal subsection of representing enough smooth colored region.
17. as the method for claim 11, wherein this source video signal comprises the signal subsection of expression I frame and implements peaking with the I frame rate.
18., wherein in level and vertical direction, all implement peaking as the method for claim 11.
19. as the method for claim 11, wherein the step of peaking adopts the high pass filter that is set to preset frequency, and the increase in the selection of this vision signal part partly has the frequency above this preset frequency.
20. as the method for claim 11, wherein the step of peaking comprises and makes this vision signal by having the band pass filter of predetermined frequency band, and the increase in the selection of this vision signal part partly falls in the predetermined frequency band.
21. as the method for claim 12, it is proportional that wherein the amplitude of the degree of peaking and frequency coefficient exceeds the scope of predetermined threshold level.
22. method as claim 11, this predetermined threshold level or spectrum distribution wherein are set,, and the selection detail section of vision signal implemented peaking and the chrominance section of the relatively flat of this signal is not implemented peaking whether corresponding to the selection level of display color details with the selection part of determining this vision signal.
23., wherein only peaking is partly carried out in the brightness of this vision signal as the method for claim 11.
24. signal processor, be used for selectively peaking vision signal in response to the frequency coefficient of the vision signal that comprises brightness part and chrominance section, this signal processor comprise importation, output and be coupling in the importation and output between processing unit, this processing unit is configured and is suitable for: receive the colourity frequency coefficient corresponding to the block of video data of the selection part of vision signal; Obtain the spectrum distribution of the selection frequency coefficient partly of vision signal; Determine the amplitude of frequency coefficient; With the amplitude of frequency coefficient and predetermined threshold level compares or spectrum distribution and predetermined spectrum distributed compare; Equal or exceed predetermined threshold level or spectrum distribution with frequency coefficient and mate this predetermined spectrum when distributing in the selection of this vision signal part, output signal, this signal indication:, should carry out the brightness part of this signal and/or the peaking of chrominance section for the selection part of this vision signal.
25. as the signal processor of claim 24, wherein processing unit is constructed to handle the signal that comprises corresponding to the segmentation of I frame, these segmentations will show with the I frame rate, and this processor is constructed to implement peaking with the I frame rate.
26., wherein only peaking is partly carried out in the brightness of this vision signal as the signal processor of claim 24.
27. a television set comprises:
Shell is equipped with signal input part, source signal processor, peaking signal processor, peaking filtering system and video display in the above;
The source signal processor, be coupled to signal input part and be coupled to the peaking signal processor and the peaking system, this source signal processor is constructed to send to the peaking system from signal input part received signal and the colourity frequency coefficient that sends corresponding to vision signal to the peaking signal processor and with vision signal;
The peaking signal processor, be coupled to the source signal processor on the input and on output, be coupled to the peaking system, this peaking signal processor is constructed to from source signal processor received signal and obtains colourity frequency coefficient corresponding to the selection part of this vision signal, the amplitude of colourity frequency coefficient and predetermined threshold level or spectrum distribution are compared, and when amplitude equals or exceeds this thresholding or mates this spectrum distribution, the output control signal is given the peaking system, so that this peaking system carries out the peaking operation;
The peaking system, be constructed to from then on source signal processor receiving video signals and receive control signal from the peaking signal processor, and come this vision signal of peaking, and will select the vision signal of peaking to send to display according to being included in from the instruction in the control signal of peaking signal processor;
Display is constructed to receive in the peaking system from then on the vision signal of selecting peaking and shows this vision signal.
28. as the television set of claim 27, wherein this source signal processor is constructed to digital video signal and decoding and this signal of decompression of received code compression.
29. as the television set of claim 27, wherein this peaking system comprises the heterogeneous mode filter of sampling again that is constructed to receive a plurality of filter taps.
30. as the television set of claim 27, wherein the source signal processor comprises analog-to-digital converter and is constructed to derive the signal frequency-domain coefficient that utilizes described converter conversion and described frequency coefficient is offered the peaking signal processor.
31. be used for a kind of method of peaking vision signal, comprise:
Receiving video signals;
Acquisition is corresponding to the colourity frequency coefficient of the selection part of this signal;
The amplitude of frequency coefficient and predetermined threshold level or spectrum distribution are compared; With
When the frequency coefficient of this vision signal equals or exceeds this predetermined threshold level or mates this spectrum distribution, the selection part of this vision signal of peaking.
32. as the method for claim 31, wherein the vision signal of Jie Shouing is an analog signal, and this signal is digital signal by parallel transformation, and obtains frequency coefficient for dividing corresponding to the selection portion in the digital signal of the selection part of this analog signal.
33. as the method for claim 31, wherein utilize digital filter to carry out peaking, this analog signal is transformed to analog signal after being transformed to digital signal before the peaking and carrying out peaking.
34., wherein in level and vertical direction, all carry out peaking as the method for claim 31.
Applications Claiming Priority (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US46640399A | 1999-12-17 | 1999-12-17 | |
| US09/466,403 | 1999-12-17 | ||
| US09/613,230 | 2000-07-10 | ||
| US09/613,230 US6810082B1 (en) | 1999-12-17 | 2000-07-10 | Chroma based adaptive signal peaking |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1347616A CN1347616A (en) | 2002-05-01 |
| CN1248486C true CN1248486C (en) | 2006-03-29 |
Family
ID=27041645
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 00806349 Expired - Fee Related CN1248486C (en) | 1999-12-17 | 2000-12-14 | Signal peaking |
Country Status (4)
| Country | Link |
|---|---|
| EP (1) | EP1157543A1 (en) |
| JP (1) | JP2003517785A (en) |
| CN (1) | CN1248486C (en) |
| WO (1) | WO2001045389A1 (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7686976B2 (en) | 2003-01-29 | 2010-03-30 | Molycorp Minerals, Llc | Composition for removing arsenic from aqueous streams |
| US8066874B2 (en) | 2006-12-28 | 2011-11-29 | Molycorp Minerals, Llc | Apparatus for treating a flow of an aqueous solution containing arsenic |
| US8252087B2 (en) | 2007-10-31 | 2012-08-28 | Molycorp Minerals, Llc | Process and apparatus for treating a gas containing a contaminant |
| US8349764B2 (en) | 2007-10-31 | 2013-01-08 | Molycorp Minerals, Llc | Composition for treating a fluid |
| US9233863B2 (en) | 2011-04-13 | 2016-01-12 | Molycorp Minerals, Llc | Rare earth removal of hydrated and hydroxyl species |
Families Citing this family (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6810082B1 (en) * | 1999-12-17 | 2004-10-26 | Koninklijke Philips Electronics N.V. | Chroma based adaptive signal peaking |
| US7161633B2 (en) * | 2001-01-10 | 2007-01-09 | Koninklijke Philips Electronics N.V. | Apparatus and method for providing a usefulness metric based on coding information for video enhancement |
| US20030080981A1 (en) * | 2001-10-26 | 2003-05-01 | Koninklijke Philips Electronics N.V. | Polyphase filter combining vertical peaking and scaling in pixel-processing arrangement |
| US6873657B2 (en) | 2001-12-27 | 2005-03-29 | Koninklijke Philips Electronics, N.V. | Method of and system for improving temporal consistency in sharpness enhancement for a video signal |
| US6862372B2 (en) * | 2001-12-27 | 2005-03-01 | Koninklijke Philips Electronics N.V. | System for and method of sharpness enhancement using coding information and local spatial features |
| US7031388B2 (en) * | 2002-05-06 | 2006-04-18 | Koninklijke Philips Electronics N.V. | System for and method of sharpness enhancement for coded digital video |
| US7054497B2 (en) | 2002-06-14 | 2006-05-30 | Koninklijke Philips Electronics N.V. | Method and system for optimizing image sharpness during coding and image enhancement |
| GB0228556D0 (en) * | 2002-12-06 | 2003-01-15 | British Telecomm | Video quality measurement |
| US7817873B2 (en) * | 2005-11-09 | 2010-10-19 | Intel Corporation | Enhancing contrast of video data while preserving sharpness |
| JP2007193397A (en) * | 2006-01-17 | 2007-08-02 | Sharp Corp | Image processing method and device for scaler and edge enhancement |
| JP5114494B2 (en) * | 2006-11-28 | 2013-01-09 | インテル・コーポレーション | Contrast enhancement of video data with sharpness |
| CN103745443B (en) * | 2014-01-10 | 2017-06-13 | 北京优纳科技有限公司 | The method and apparatus for improving picture quality |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5454051A (en) * | 1991-08-05 | 1995-09-26 | Eastman Kodak Company | Method of reducing block artifacts created by block transform compression algorithms |
| US5168375A (en) * | 1991-09-18 | 1992-12-01 | Polaroid Corporation | Image reconstruction by use of discrete cosine and related transforms |
| EP0585573A3 (en) * | 1992-08-31 | 1994-07-13 | Ibm | System and method for suppressing blocking artifacts in decoded transform coded images |
| US5515112A (en) * | 1994-12-19 | 1996-05-07 | Tektronix, Inc. | Chrominance transition corrector |
| EP0721286A3 (en) * | 1995-01-09 | 2000-07-26 | Matsushita Electric Industrial Co., Ltd. | Video signal decoding apparatus with artifact reduction |
-
2000
- 2000-12-14 EP EP00983326A patent/EP1157543A1/en not_active Withdrawn
- 2000-12-14 WO PCT/EP2000/012731 patent/WO2001045389A1/en active IP Right Grant
- 2000-12-14 JP JP2001546150A patent/JP2003517785A/en active Pending
- 2000-12-14 CN CN 00806349 patent/CN1248486C/en not_active Expired - Fee Related
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7686976B2 (en) | 2003-01-29 | 2010-03-30 | Molycorp Minerals, Llc | Composition for removing arsenic from aqueous streams |
| US8475658B2 (en) | 2003-01-29 | 2013-07-02 | Molycorp Minerals, Llc | Water purification device for arsenic removal |
| US8066874B2 (en) | 2006-12-28 | 2011-11-29 | Molycorp Minerals, Llc | Apparatus for treating a flow of an aqueous solution containing arsenic |
| US8252087B2 (en) | 2007-10-31 | 2012-08-28 | Molycorp Minerals, Llc | Process and apparatus for treating a gas containing a contaminant |
| US8349764B2 (en) | 2007-10-31 | 2013-01-08 | Molycorp Minerals, Llc | Composition for treating a fluid |
| US8557730B2 (en) | 2007-10-31 | 2013-10-15 | Molycorp Minerals, Llc | Composition and process for making the composition |
| US9233863B2 (en) | 2011-04-13 | 2016-01-12 | Molycorp Minerals, Llc | Rare earth removal of hydrated and hydroxyl species |
Also Published As
| Publication number | Publication date |
|---|---|
| EP1157543A1 (en) | 2001-11-28 |
| WO2001045389A1 (en) | 2001-06-21 |
| JP2003517785A (en) | 2003-05-27 |
| CN1347616A (en) | 2002-05-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1248486C (en) | Signal peaking | |
| US11064209B2 (en) | System and method for content adaptive clipping | |
| US8243797B2 (en) | Regions of interest for quality adjustments | |
| CN1110961C (en) | Post-filter for removing ringing artifacts of DCT coding | |
| CN1130918C (en) | Picture signal processing method and apparatus | |
| US8750390B2 (en) | Filtering and dithering as pre-processing before encoding | |
| JP4717138B2 (en) | Image encoding method and image encoding apparatus | |
| US9313504B2 (en) | Pattern detection module with region detection, video encoding system and method for use therewith | |
| CN1665298A (en) | Method of removing noise from digital moving picture data | |
| CN1267817C (en) | Signal indicator for fading compensation | |
| EP2782346A1 (en) | Quantization matrix design for HEVC standard | |
| US20100315558A1 (en) | Content adaptive noise reduction filtering for image signals | |
| US6903782B2 (en) | System and method for performing segmentation-based enhancements of a video image | |
| CN1347620A (en) | Method and architecture for converting MPE G-2 4:2:2-profile bitstreams into main-profile bitstreams | |
| CN1695381A (en) | Sharpness enhancement in post-processing of digital video signals using coding information and local spatial features | |
| US8145006B2 (en) | Image processing apparatus and image processing method capable of reducing an increase in coding distortion due to sharpening | |
| US20050024651A1 (en) | Adaptive complexity scalable post-processing method | |
| US6810082B1 (en) | Chroma based adaptive signal peaking | |
| US8401322B2 (en) | Image recording device, image recording method and program | |
| US8442338B2 (en) | Visually optimized quantization | |
| US10034004B2 (en) | Receiving apparatus, transmitting apparatus, and image transmitting method | |
| CN102316249B (en) | Image processor, display device and image processing method | |
| CN1801947A (en) | Method and apparatus for encoding and decoding moving images, and its recording medium | |
| JP6200220B2 (en) | Image processing apparatus, encoding apparatus, decoding apparatus, and program | |
| WO2010101292A1 (en) | Method of and apparatus for processing a video image |
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 | ||
| C19 | Lapse of patent right due to non-payment of the annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |