JPS58105678A - Image signal processing method - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for processing image signals such as high-definition television signals, and in particular, to improve image quality and display high-definition images without increasing the required frequency bandwidth. This is how it was done.

In this type of conventional signal processing method, Ijii signal processing is used to reduce the required transmission frequency bandwidth without degrading the quality of the displayed image on the receiving side of the image signal transmission system. Generally, so-called line interlaced scanning is used. However, WJJ1 displayed using normal line interlaced scanning
In FLL, the coarse scanning line structure of each field is easily noticeable and is a major cause of deterioration in image quality.In addition, for images with movement, the quality deteriorates due to the loss of the naturalness of the movement of the image bond. There were various drawbacks, such as the occurrence of In other words, the line interlaced scanning method iii#I, which has an alternating scanning line structure for each field, requires about 70% of the vertical resolution expected from the number of scanning lines.
Although there is an essential drawback that only vertical solutions can be obtained,
It has the characteristic that flicker can be reduced without increasing the number of waves per second, therefore, the illumination wave number bandwidth can be reduced without increasing the wave number bandwidth. Even if an image is constructed using a line sequential scanning method that doubles the number of images per second, the number of images per second must be increased in order to adapt to the line interlaced scanning method during playback and to avoid an increase in the transmission frequency bandwidth. It was necessary to intermittently sample the doubled line-sequential scanning image signal, perform time axis expansion, and convert the image scanning method.

Therefore, II! i
When sampling is applied to the i-image signal, in addition to the frequency components associated with image scanning, the sampling frequency and its sideband wavenumber components are newly generated, changing the distribution of the original image signal components.
In particular, the essential problem of image quality deterioration associated with sampling is that the quality of the high-definition image signal is significantly degraded due to the high frequency band being mixed into the distribution of high-frequency signals. This is a particularly serious problem for the processing method of jl's signals. Conventionally, in order to remove unnecessary signal components mixed into the III image signal due to such sampling, a simple configuration P that simply deals with the arrangement of each signal component on the frequency axis has been proposed.
However, in order to reproduce and display high-quality LKI, it is necessary to arrange each signal component not only on both the horizontal and vertical frequency axes, but also on the time axis, that is, the *I signal spectrum. It is necessary to perform image signal processing that removes unnecessary signal components mixed in based on sampling of 1 gI signal from all angles using a corresponding P-wave device configured in two or three dimensions. .

In addition, when converting the scanning method mentioned above, the conversion of the scanning line order in the transverse direction and the time axis conversion are carried out exclusively, so the case of converting the scanning method between sequential scanning and dot interlacing etc. Except, mainly,
It is common to use a p-wave device configured in λ dimensions with respect to the vertical frequency axis and the time axis.

SUMMARY OF THE INVENTION An object of the present invention is to provide an image quality signal processing method that does not cause image quality deterioration inherent in interlaced scanning when reproduced and displayed using interlaced scanning.

Another object of the present invention is to remove unnecessary signal components in the P range corresponding to the lfif* signal spectrum on a multi-dimensional coordinate plane consisting of at least a vertical frequency axis and a time axis.
An object of the present invention is to provide an image signal processing method using a wave generator.

Still another object of the present invention is to provide an image flaw signal processing method that further improves image quality by strengthening Takagi signal components in the passband of the multidimensional F-wave filter used. Still another object of the present invention is to remove unnecessary signal components that occur when sampling an image signal using a multidimensional P-wave signal on at least one of the transmitting side and the receiving side in an image signal transmission system. I did it! The purpose of the present invention is to provide an iW signal processing method.

That is, the image signal processing method of the present invention performs sampling of the flaw signal in multidimensional coordinates consisting of a coordinate plane formed by at least the vertical frequency axis and the time axis among the horizontal frequency axis, vertical frequency axis, and time axis regarding the image signal. At least a vertical frequency axis and time generated by desiring a region on the origin side of the multidimensional coordinates of a line connecting at least seven sampling frequency points generated based on the origin and points that are approximately equidistant from the origin of the multidimensional coordinates, respectively. This is characterized in that unnecessary high-frequency signal components on the @ coordinate plane formed by the axes are suppressed.

The invention will be described in detail below by way of example with reference to the drawings #i
to develop.

First, FIG. 7 shows an example of the basic configuration of an image signal processing apparatus according to the present invention. The basic configuration shown in the figure will be described in detail later, but what is in accordance with the above-mentioned characteristics of the present invention is the multi-dimensional prefilter λ and interpolation filter 7b inserted in the transmitting side and the receiving side, respectively. Among them, the multidimensional prefilter λ is the subsampling circuit 3.
Before the image signal that is intermittently extracted is expanded in the time axis by the time axis conversion circuit to reduce the signal frequency band, the sampling frequency component newly generated by sampling and its sideband components are converted to virtual sampling frequency points. This is to prevent the signal from becoming a so-called folded signal component, which is the origin of the signal, and being mixed into the original image signal spectrum.

That is, a broadband image signal (a) of, for example, 521 scanning lines and 10 frames per second, obtained from a high-definition television camera using a line-sequential scanning system, generally has a horizontal and vertical frequency axis and a time axis. On the vertical frequency/time coordinate plane in three-dimensional coordinates as shown in Figure 1,
The signal spectrum distribution of the subject itself within the area surrounded by the dotted line, which has a change in the vertical direction of 1fiJ with the origin 0 as the center, and the spread according to the movement, and the image f* of the subject with this signal spectrum taken by the camera l. On the vertical frequency axis fv, the point O (/y□ + O) determined by the vertical scanning wave number f, and the point (0, /, ) and the point (/y y□ + /
Each of these exhibits a spatial spectrum consisting of a signal spectrum generated in a region having a spectral spread of seconds similar to that of Kei described above, centering on F). In addition, tIp1
In FIG. 1, the frequency point /VO on the vertical frequency axis and the frequency point f1 on the time axis are both normalized and shown equidistant from the origin.

A broadband image signal exhibiting such a spatial spectrum (~t
, as shown in FIG. 7, directly lead to the sub-sampling circuit 3 and the time axis conversion circuit 3 without going through the multi-dimensional prefilter λ according to the present invention, for example, the scanning AT column,
When converting to a normal narrowband image signal (&) using a 60-field, 30-frame interlacing method, the sub-sampling circuit 3 extracts odd and even lines for each frame from the wideband image signal (&). Assuming that the time axis of the intermittent image signal Φ) extracted alternately is converted to double the time axis to form a normal narrow band -fil signal, the space of the intermittent image signal Φ) extracted from the sub-sampling circuit J The spectrum is created by a new coordinate point (/) determined by the new vertical scanning wave number /vo/co and the number of images per second /F/co based on the signal launch in the area mentioned above for the wJ2 diagram.
At least a new signal frequency spectral component in the area shown by the solid line in Figure 1, which has the same spread as above and centered on VO/Co /F/Co), is newly generated, and as shown in Figure 2, An overlap occurs between the original signal spectral distribution regions surrounded by dotted lines. Of these, the nine overlaps that occur between the original signal spectrum distribution region centered on the coordinate origin O are at the point ('vo/2 + 'F/, 2)
is mixed into the original signal spectrum distribution as a so-called Su/Grinder aliasing component with the virtual origin 08, causing image quality deterioration due to aliasing distortion in the image signal.

The present invention eliminates the above-mentioned aliasing distortion components that occur based on signal sampling during conversion processing of this kind of 1IIR signal, at least those that occur on the λ-dimensional coordinate plane regarding the vertical frequency axis and the time axis. , it is possible to reproduce an image of sufficiently high quality.For example, the sampling aliasing component mentioned above in FIG. 2 spreads around the temporary S origin 08 and invades the original signal spectrum distribution region. Because,
In order to avoid overlap between the original signal spectrum distribution region centered on the origin 0 and the folded component region centered on the virtual origin 08, a point located at a distance from both the origin 0 and the virtual origin 08, that is, A point on the vertical cylinder wavenumber axis (/vo//
2. o > and a point on the time axis (o.

After limiting the original signal spectral distribution area to only the area closer to the origin 0, the straight line connecting the 0 and 0 lines, intermittent extraction and time axis conversion are performed with the number of scanning lines set to - as described above. is applied to the wideband image signal (a), and the original signal spectral distribution area is restricted using a two-dimensional or three-dimensional low-pass (pass) P-wave device. .

However, according to the present invention, the original signal spectral distribution area of the wideband image signal itself is reduced to some extent, but the signal components in the reduced area are mixed with sampling aliasing components. Compared to the reproduced image quality when the signal is returned to its original state with a punch, the high-frequency signal components are somewhat lost, but the image quality obtained when sampling aliasing distortion components are removed is much better. It has excellent performance and can reproduce and display desired high-quality images.

That is, in the image signal processing apparatus according to the present invention shown in FIG. The above-mentioned points ('VO/J,
O) Oyohi (0, IF/, 2) After removing the signal component of the diagonally shaded part outside of the straight line passing through wo,
The sub-sampling circuit 3 and the time axis conversion circuit Hiroshi perform the above-described intermittent extraction of the scanning line and time axis expansion to form a normal narrow-range picture signal (0), and then use the modulation device! On the receiving side, the narrowband image signal extracted by the demodulator t is guided to the time axis conversion circuit 7a to convert the time axis to the original wideband image. A line-sequential scanning type pseudo broadband image signal restored to the state of signal (a) is generated, and a high-quality image is reproduced and displayed on the display device l. However, in the narrowband 111ii signal extracted on the receiving side, the original signal spectral distribution area of 10 is limited by the prefilter on the transmitting side, so the above-mentioned aliasing 17 distortion component does not mix. However, due to intermittent extraction of scanning lines in the sub-sampling circuit 3, points (2vO, O), (0,
/F), there may be cases where unnecessary signal components based on sampling remain in contact with the limited original signal spectrum distribution area, and new signal components based on sampling may remain outside the straight line connecting Since undesired JLI signal components are generated, these unnecessary signal components are also displayed as noise on a display device having a wide band characteristic for high-quality JLI display at Hz, degrading the display image quality. Therefore, in the image signal processing apparatus according to the present invention shown in FIG. The signal is supplied to the display device l to prevent noise components due to sampling or the like from being displayed.

Next, we will explain the above-mentioned image signal processing apparatus using the multi-dimensional low-pass P-wave device according to the present invention, and with reference to FIGS. For example, the number of scanning lines jt2 from l! ,
Assuming that the line-sequential scanning broadband image signal (a) of the tO frame has the signal waveform shown in waveform (a) in Figure 3, in the case of a stationary 1i1ii run with no movement of the ijii, its The signal frequency spectrum is shown in the waveform shown in Fig.
As shown in ~, when the horizontal scanning frequency is C/H, horizontal frequency axis spectral components are arranged at frequency positions that are integral multiples of the horizontal scanning frequency, and vertical frequency axis spectral components are arranged at frequency positions that are integral multiples of the frame frequency fv before and after that. When a part of the signal frequency spectrum is enlarged, it becomes a waveform Φ) in Fig. Φ. That is, the horizontal scanning frequency in the wideband image signal (a) in the illustrated example is "fH", which is one time the horizontal scanning frequency fH in the normal television system, and the arbitrary horizontal frequency axis spectral component n/H (n - / .

) on both sides of the frame frequency fvm 40
HzOI! Vertical frequency axis spectral components as sideband components are arranged at one wave number interval. When the wideband s1m signal (a) exhibiting such a frequency spectrum is supplied to a device filter consisting of a -dimensional to three-dimensional low-pass filter, its prefilter λ is shifted to a point (//) in the original signal spectrum distribution region as described above. vo/yu 0),
(0, /F/J), so as shown in waveform (0) in Figure 1, the horizontal frequency axis in the frequency spectrum shown in waveform (b) In the arrangement of the vertical frequency axis components centered on the component λn/a, a part of the high frequency band is deleted and horizontal passbands with nine center frequencies are obtained. As shown in the wideband image signal (~ is guided to the sub-sampling circuit J and shown in the waveform Φ in FIG. 3) that has passed through the pre-filter with such a pass characteristic, one high frequency signal is generated every other horizontal scanning period of the sequential scan. When a signal is extracted and subjected to signal thinning, that is, so-called subsampling, the entire band ri remains almost unchanged, but the horizontal fundamental frequency decreases by -λ. Next, refer to the time axis conversion circuit.
As shown in waveform (0) in Fig. 3, when the extracted signal component is expanded by a factor of t- times between the axes, the signal band is halved, and the entire band becomes -, resulting in a scan corresponding to λ in the standard television system. It corresponds to the interlaced narrow-band clay signal with 511 lines, 4o fields, and 30 frames. That is, the wideband rIiii signal spectrum shown in waveform (b) in Figure 1 is divided into horizontal frequency axis components at frequency intervals of the horizontal scanning frequency fH, as shown in waveform (d) in Figure 1. are arranged, and the vertical frequency axis components extend opposite to each other on both sides of the horizontal frequency axis components at an interval of a field frequency of 40 Hz. Therefore, if the broadband image signal spectrum of waveform (in the waveform) is sampled and time-axis converted in the form of t, as shown by the thin line in waveform (d),
The vertical frequency axis components extending opposite to each other from the horizontal frequency axis components kon/H and (Jn+t)/a intersect with each other and mix with each other, for example, □Noise due to distortion components corresponding to the sampling aliasing distortion components mentioned above in Fig. 1, such as a vertical frequency axis component of JQHz occurring in relation to the horizontal frequency axis component, is
This will cause the image quality to deteriorate. However, with a prefilter having a passband characteristic as shown in waveform (0), the spectrum of the central portion between adjacent horizontal frequency axis components in the nine wideband III signal frequency spectrum shown in waveform Φ) If these components are removed in advance, as shown by the thick line in the waveform (b), there will be no mixing due to the intersection of the vertical frequency axis components, and therefore, equivalently, the high The quality image signal can be reproduced and displayed, and the first
In the figure, this corresponds to limiting the vertical spatial frequency component to /vo/yu or less.

In the above explanation, the wideband image signal (~) is assumed to be a static signal, but if the wideband image signal (4) represents a moving image, the wideband image signal (4) represents a moving image.
The position of the image shifts for each successive frame of the image signal (a), and the time interval required to reproduce the same pixel differs from the vertical scanning cycle length, etc., so the waveform shown in Figure 3) is A spread occurs in the frequency position of the vertical frequency axis component of the to H2 interval in the signal frequency spectrum, and when the image moves rapidly, the spread increases and the above-mentioned intersection of the vertical frequency axis components between adjacent horizontal frequency axis components occurs. Cross-contamination similar to contamination also occurs between adjacent vertical frequency axis components that have a certain degree of spread, and the movement of the reproduced and displayed image loses its naturalness, resulting in motion iijigIIi! i
This will cause deterioration in quality.

However, a wideband moving image signal having a frequency spectrum exhibiting the spread of vertical frequency axis components (→ is the frequency interval 40 shown in the waveform (0) as described above)
When supplied to a pre-filter with a comb-shaped passband characteristic of Hz, high-frequency components that cross-mix with each other in the spread of vertical frequency axis components are removed. low range P
#! When performing 1i1i1i signal processing using a device,
It is possible to effectively prevent the conventional image quality deterioration, and it is congratulatory that the spatial frequency component t /F/ in the time axis direction is limited to the following in FIG. 2. Therefore, the effect of an F-wave generator exhibiting a comb-shaped passband characteristic as shown in waveform (0) in Figure 1 of the frame wave number interval on the frequency axis is as follows.
The point (jVo//2. O>, (0.

/F/, ) corresponds to the effect of the restriction on the straight line passing through.

In the above description, for example, the number of scanning lines is 121.
, line-sequential scanning wideband with to frames per second - ** number of scanning lines jλ! , to fields per second, JO frame interlace method narrow band The aspect of image signal processing according to the method of the present invention has been described in the case of converting it into an image signal of the method, but similarly, 3
When converting to a high-order interlaced image signal such as ;l, Fis', i, etc., aii[signal is
Extraction 1. Then, the same image processing described above will be applied. Furthermore, when converting to a dot interlaced image signal, the line-sequential scanning broadband image signal is sampled at a much higher sampling frequency, for example several hundred times higher than the horizontal scanning frequency; The image signal is intermittently extracted once per λ sample or once every J samples in the same way as described above in FIG. It will be processed. Note that as the multidimensional low-pass filter used in such a case, a three-dimensional low-pass p-wave filter based on three-dimensional coordinate axes including the horizontal frequency axis is used, as described above.

Next, FIG. 1 shows an example of a specific configuration of the time axis conversion circuit in the basic configuration shown in FIG. 1 of the image signal processing apparatus according to the present invention. In the time axis conversion circuit having the configuration shown in the figure, a broadband image signal (b) which is intermittently extracted as shown in the waveform (b) of FIG. 3 is supplied to the input terminal, and an image 111 signal digital signal is supplied to the input terminal Provides a sampling clock signal for sampling. The intermittent extracted signal from the input terminal 21 is introduced into a low-pass P-wave filter/It/C to remove unnecessary signal components outside the required band, and then an analog signal driven by the clock signal from the input terminal 21 is introduced. The intermittent digital image signal is led to a digital converter 815// to be converted into an intermittent digital image signal L, and the intermittent digital image signal is alternately supplied to the 1H34 extension lines 13 and 14I, each having a storage capacity for l scanning lines, via a changeover switch /Jt. , is driven by a clock signal applied from an input terminal λl via a changeover switch 1, and is written alternately. These /H delay groups 13 and /D also have an input changeover switch λQ.
The image signals for l horizontal scanning period, which were written alternately as described above, are read out alternately with the time axis expanded to 1 times, and the 1! J.I.
A narrowband image signal (0) having the form shown in waveform (0) in the JJ diagram
) and lead it to the digital-to-analog converter 14,
The converted output analog narrowband image signal is subjected to low-pass wave@/7 to remove unnecessary signal components outside the desired band.1. At Tatsue, take out the time axis conversion output from the output terminal /l.

In addition, each changeover switch saw, /j, /ri, koO
are switched in conjunction with each other. When performing high-speed writing of an image signal to one of the 'H delay lines /J, /' in response to an input signal, the 'H delay lines /J, /' are switched in conjunction with each other.
It is assumed that the image key signal is read out at low speed from the other side of 4C in response to a low-speed clock signal.

Note that in the configuration shown in the first diagram, if the wideband image signal supplied to the input terminal has already been digitized, the analog-to-digital converter l in the configuration shown in the diagram
Of course, if the subsequent signal processing is carried out in digital form, the digital Φ-to-analog converter/4 in the illustrated configuration can be omitted.

The converted output ms multiplied signal 0) shown in the waveform (0) in Fig. 3 obtained as described above from the time axis conversion output reference in the basic configuration shown in Fig. 1 is an interlaced signal obtained from a normal television camera. It has the same signal form as the image signal, but before the signal conversion is performed, a pre-filter is used to apply a spatial and temporal three-dimensional wave, and as a result of the signal conversion, the original signal spectrum distribution is This method is significantly different from conventional interlaced image signals in that it prevents the occurrence of so-called sampling aliasing distortion components mixed into the image signals.

Next, we will discuss the F-wave characteristics of the three-dimensional low-pass (pass) P-wave device, which forms the singularity of the flaw signal processing method according to the present invention, and its Pff.
A specific configuration for realizing % property will be explained.

First, in order to define the three-dimensional one-wave characteristics, it is necessary to
As shown in Figure 1, the image of the town is spatially sampled in λ dimensions in both the horizontal and vertical directions by horizontal and vertical screen scanning, and temporally as well. Analyze by approximating what is being sampled. Therefore, the sampling frequency in spatial one-dimensional sampling corresponds to the horizontal and fl[ scanning frequencies in screen scanning. For example, in the line sequential scanning method, the vertical sampling frequency fv0 is This corresponds to the frame frequency that is the escapism of the number of lines Ln, and Jvo-'/
Ln, and the horizontal sampling frequency fu
o is based on the sampling of the IgI signal in the horizontal direction when converting an analog image signal to a digital IgI signal, and the standard method is calculated by dividing the sampling frequency by the horizontal scanning frequency fhK. In television, the value multiplied by the aspect ratio of the screen, which is 3; that is, 4, is expressed in the spatial frequency form 11,
For example, in the scanning method, the temporal change due to the motion K of iiijgI appears on a frame-by-frame basis, so the number of frames per second is 1N. When the sampling frequency of each axis in the three-dimensional coordinate axes for defining the three-dimensional P wave characteristics is set in this way, the image signal of the line sequential scanning method is expressed in the spatial frequency/time domain represented by the three-dimensional coordinates. The resulting signal spectral distribution can be expressed using a discrete-Fourier transform, or so-called DFT method. Also, for interlaced image signals, the interlacing itself can be expressed using a type of sampling different from that in the line-sequential scanning method. If we consider that there is,
Similarly, it can be expressed using the DFT method, and in that case,
The image signal spectral distribution in the above-mentioned three-dimensional coordinate area can be determined in the same way for the sampling frequency in each direction, for example, for the monochrome and interlaced methods, and also for the It is possible to estimate whether a distortion component will occur, and furthermore, based on the estimation, it is also possible to clarify the image signal spectral distribution characteristics necessary to prevent the mixing of aliasing distortion components.

17 As shown in Fig. 7, each sampling frequency of the line sequential scanning method is expressed in three-dimensional coordinates consisting of the horizontal and vertical spatial frequency axes and the time axis is the frequency on each axis. Point/uo, jVo, ft.

and their respective harmonic frequency points J/J/uo,
vo.

They are set at the positions of the black dots in the figure, which are determined by the combination of ko and lo.

However, as described above with reference to FIG. 2, in general, sampling is not performed in the horizontal spatial frequency axis direction in image signals, and between the line sequential scanning method and the interlaced method, sampling is not performed in the horizontal spatial frequency axis direction. Since there is no change in Signal spectral distribution can be considered.

Each sampling frequency of the interlaced image signal expressed in such λ-dimensional coordinates is shown in FIG. 1 and the set of each harmonic frequency point/j thereof.

They are set at the positions of the black dots shown in the figure, which are determined by the alignment.

As is clear from a comparison between FIG. 7 and FIG. 1, in the interlaced jIgI signal, the frequency points on each axis are
Therefore, as described above with reference to FIG. If the signal spectrum distribution of the line-sequential scanning wideband j1gN signal is restricted to the boundary line that is the distance distance, that is, the area on the origin 0 side from the straight line passing through on each axis, then the λ:l interlaced narrowband image g
When converted to 1 signal, the original signal is smooth! It is possible to prevent the occurrence of image quality deterioration due to the mixing of sampling aliased components into the torque distribution region, and the resulting signal spectrum distribution limited region is shown with diagonal lines in FIG.

Therefore, Fig. 1 shows the signal spectral distribution limited area when converting a line-sequential scanning wideband single signal into an interlacing narrowband single signal. The virtual origin Os is shown as 082 to distinguish from the case. In addition, Figure R shows only the first quadrant of the coordinate plane regarding the vertical frequency axis and time axis, and if the other quadrants are combined, the signal spectrum distribution restricted area in this case is indicated by the diagonal lines shown in the figure. It becomes a square made up of 1ft4c triangles. Therefore, the distribution area of the aliased component spectrum centered on the virtual origin 082 can also be roughly represented by a rectangle of substantially the same shape as the visually diamond-shaped area shown in the first diagram.

Therefore, in order to realize an a-dimensional low-pass F# filter on the vertical frequency axis and the time axis whose pass band is such a signal spectrum distribution limited area, it is necessary to equivalently use the area shown with diagonal lines in Figure (a). A two-dimensional low-pass F-wave device with a passband of
Shown below. The one-dimensional one-wavelength generator with the illustrated configuration includes a low-frequency f-wave generator in the time axis direction and a low-frequency p-wave generator in the vertical frequency axis direction. are combined by adders DD1 and DD2, and the low-frequency P wave unit in the time axis direction is a shaded area with a diagonal shaded area whose upper limit is point α on the time axis Jt in FIG. 2(a). is the passband, and the vertical frequency axial low-frequency waveform generator λj that subtracts and supplies the passband component from the input image signal supplied to the input terminal , 2J has the passband of the second FA ( If the hatched area with α as the lower limit is defined as the pass band, and the filtered output components of the -wavelength devices C and C are added together and taken out from the output terminal T, the entire shaded area in Fig. A λ-dimensional wave output III signal with the region as a passband is obtained.

In addition, in FIG. 9(a), the passband shown in FIG. 1 is approximated in a relatively simple form, but in order to approximate this more accurately and precisely, the diagonal line shown in FIG. 1O(a) is used. A one-dimensional filter using a shaded area as a passband can be constructed as shown in Figure 1). The one-dimensional one-wave tatami with the illustrated configuration has diagonal shaded areas whose upper limits are points α and α on the time axis, in the same way as in the λ-dimensional Pa1e with the configuration shown in FIG. 2(b). In the vertical frequency axis direction, the passbands of the low-pass filters 1 and λ are defined as their respective passbands, and the diagonal shaded areas whose upper limits are points β and β2 on the vertical frequency axis. There are nine combinations of the low-frequency PflI filter and the adders MuDD5 to MuDD5.

Next, λ shown in Fig. 2(b) & io)
An example of a specific configuration of a low-pass filter in the time axis direction or vertical frequency direction used in the configuration of a one-dimensional one-wave filter is shown in Figure 1/0. A plurality of connected delay elements 3≠, those delay elements 34! - a so-called transversal filter consisting of a plurality of weighting coefficient units 3j and an adder DDd that adds the output signals of the coefficient units 3j, and each delay element J Reference is made to the Fi/frame memory in the low frequency waveform device in the time axis direction, and the l line memory in the low frequency P waveform device in the direct wavenumber axis direction.

Next, the line sequential scanning wideband iir* signal is
In the same manner as shown in FIG. 1 in the case of converting to an interlaced narrowband 1iiif& signal, the manner of limiting the signal spectral distribution region of the co-dimensional P-wave device in the case of converting to a J:/interlaced narrowband 1iiif& signal will be described in the following. This is shown in Figures (4) and (B). As is clear from the figure, the sampling frequency point newly generated when converting to the J2l interlace method 1iii signal is increased by 3jll to 11 determined by the interlace ratio J:l. Therefore, J2/interlaced system ilii
For the signal spectral distribution limited region in the case of converting to a gII signal, the above-mentioned sampling frequency points 0□ and 0-s are set in the same manner as in the case of converting to a 22l interlaced image signal as shown in the figure. As temporary SZ points, their virtual origins 08J1. O4s and coordinate origin 0
If we subtract the boundary line tyIc that is equidistant from each other, we will get the first . Ijfi1 minute, and the distribution area of the sampling aliasing distortion component spectrum centered on each sampling frequency point 08s and OQs is as follows.
It becomes a regular octagon with 11 polygons.

Next, the manner of limiting the signal spectral distribution area of the λ-dimensional F-wave device when converting a line-sequential scanning method broadband completed signal into a reference signal of the J:l interlace method is explained below. (* When converting to
Shown in Figure J. However, as is clear from the figure, Hiroshi:/
The newly generated sampling frequency point when converting to an interlaced image signal has an interlace ratio of 4L:
A black spot appears at the position of 084 t OB4' t Os4. Therefore, ≠:l For the signal spectral distribution limited region when converting to an interlaced image signal, the above-mentioned sampling frequency point oli410
If we take 941084 as the virtual origin and find the boundary lines that are equidistant from the virtual origin 0841 ``84' l Os/ and the coordinate origin 0, we can find the coordinate origin 0 as shown by diagonal lines in Figure 1J. The distribution area of the sampling aliasing distortion component spectrum centered on each sampling frequency point 084.O84', 084 is the first quadrant of an irregular ten triangle centered on .
becomes an irregular 10-triangle that is approximately the same shape. Therefore, the desired two-dimensional low-frequency single wave filter can be realized in the same manner as described above with reference to FIGS. 2 to 1I.

Furthermore, as is clear from the above-mentioned constraints, even when the interlace ratio is further increased, new sampling frequency points close to the required band are determined by the vertical network (Fig. 73) in the same manner as described above. In the same manner as above, a desired signal spectrum distribution restriction region is determined, and a one-dimensional low-frequency wave powder with such a signal spectrum distribution control region as a pass band is configured as described above, thereby interlacing with various interlace ratios is obtained. Even in the case of conversion to a digital image signal, it is possible to prevent image quality from deteriorating due to mixing of sampling aliasing distortion components.

The above-described signal spectrum distribution restriction area for preventing sampling aliasing components from being mixed in according to the present invention is not limited to the above-mentioned line interlaced case, but also applies to converting a line sequential scanning wideband 1-element signal into a dot-interlaced 1-element signal. The present invention can be similarly applied to other cases, and examples of aspects of limiting the signal spectrum distribution area in such cases are shown in FIG. 741L and FIG. tS. That is, as in the case of converting to the line interlacing method 1 described above, successive pixels of an image constructed by line sequential scanning are divided into every other group 0.0, as shown in FIG. The sampling frequency points that are newly generated when coni/dot interlacing is performed by dividing into X are also sampled in the horizontal frequency axis direction. ; The horizontal frequency axis, vertical frequency axis, and time axis determined by l each appear at the black point h1 in the diagram, that is, at the diagonal vertex opposite the coordinate origin O in the cubic shape at) Od8fhO where each frequency is determined by , the signal spectral distribution limited area based on the same idea as in the case of converting to a line interlaced signal is the vertices a, b, o, d, e of the cube.

This is an area surrounded by line segments connecting f and O and the intersection point g of the diagonal lines. Therefore, in order to actually construct a three-dimensional low-pass p-wave device whose required pass band is the signal spectrum distribution restriction region of such a complicated shape, it is necessary to create such a complicated shape as shown in FIG. 76(a). By approximating with a relatively simple shape as shown in FIG. 1, a substantially equivalent F-wave effect can be obtained.

Therefore, in order to implement the three-dimensional low-pass Pa device shown in Fig. i4 (JL), which has a well-shaped pass band, it is necessary to use Fig. 2 (a).
The two-dimensional low-pass P-wave device having the passband shown in Fig. 1 is divided into a band in the time axis direction and a band in the vertical frequency axis direction, as shown in Figure (b). In the same manner as shown in FIG. Adder aO to adder DD
7. By combining Mu DD8 and Mu DD9, it can be easily constructed.

In order to improve the accuracy of the approximation of the passband of the three-dimensional low-pass F-wave filter with the configuration shown in Figure 14 (in Figure 14) to the required passband shown in Figure 14, As shown in FIG. The low-frequency f-wave devices 31 can be combined by an adder module DDIL module 1λ so that the resultant passband closely approximates the required passband shown in FIG.゜In the case of converting to a λ;l dot loose image signal, the configuration of the three-dimensional low frequency waveform generator described above increases the interlace ratio and converts the 9-dot signal into The present invention can be similarly applied to the case of converting to an interlaced image signal.

Therefore, in the image signal processing device according to the present invention shown in FIG. and Time Axis Conversion Circuit Teiyoha, Part 3
As described above with reference to the figure, the narrowband signal (0) converted from the sequential scanning wideband image signal (a) is a modulator! The signal is converted into a carrier signal form and transmitted, and demodulated to its original form by a demodulator t on the receiving side. In addition, the transmission cost demodulator! , 4 can be either analog or digital, and conventional devices can be used as they are.

On the receiving side, the narrowband picture signal (0) restored as described above is guided to the time axis conversion (b) path 7a, and is inversely related to the time axis conversion by the time axis conversion circuit on the transmitting side. After performing axis conversion, it is reconverted to an intermittent signal (b) by high-speed scanning as shown in Fig. 3(b), and a signal similar to that shown in Fig. 1 is generated for the transmission time axis conversion circuit. The circuit configuration performs appropriate interpolation on the intermittent image signal to convert it into a line-sequential scanning pseudo-wideband image signal. Although this pseudo-wideband one-way signal has been subjected to preventive measures against image quality deterioration due to the mixing of sampling aliasing distortion components by the prefilter λ on the transmitting side, after the preventive measures, the sub-sampling circuit J is sampled to generate a new sampling frequency component, and furthermore, the time axis conversion circuit 7a on the receiving side
Since there is a risk that unnecessary signal components adjacent to the signal spectrum distribution restricted region may be mixed in when undergoing time axis transformation by A three-dimensional low-frequency wave with a passband exactly the same as that of the three-dimensional prefilter is applied, and through such signal processing, adjacent nine-tone signal components are removed to create a high-quality line-sequential scanning wideband image signal. is supplied to the display device l so that it can display a high-definition image.

Note that the configuration of the three-dimensional filter 7b can be simplified to simplify the configuration of the receiving side device and reduce the cost, but even in such a case, the three-dimensional spatial frequency generated by sampling the 1jif# signal Low sampling frequency components in the region, i.e., for example, folded component regions centered at sampling frequency point 0.1 shown in FIG. 1, or centered at sampling frequency points 0850≦ shown in FIG. 1, respectively. Sampling aliasing components distributed in the aliasing component region are sufficiently suppressed,
It is essential to make it difficult to visually detect in public images.

Furthermore, ii! according to the method of the present invention shown in FIG. The order of the time axis conversion circuit 7a and the interpolation filter 7b in the reception unit 111 of the iif & signal processing device is changed to add unnecessary sampling frequency aFit, minutes due to the time axis conversion to the low speed scanning 111g1 signal before time axis conversion. It is also possible to apply the three-dimensional low-frequency f-wave to be removed and then perform time axis transformation. An example of the configuration of the main part in such a case is shown in FIG. 17 in the case of converting a J:/line interlace system l1g11 signal to a line sequential scanning system broadband 1jif# signal. In the illustrated configuration, a low-speed scanning interlaced narrowband image signal t as shown in FIG. 3(0) from the demodulator 4 is directly passed through a three-dimensional interpolation filter as a non-delayed image signal jj. While leading to the l-field memory shift, the l-frame delayed image signal by the tei r and -line/memory code, the l-frame delayed image signal j3 and the l field +-line delayed image signal! #iJ
While limiting the signal spectral distribution region of the f# signal as described above, and using the l field ten-line delay image signal well as a reference, the subsequent time axis conversion circuit is applied as shown in FIG. @3 figure Φ formed by /
) A rounded interpolation signal is generated in advance to supplement the intermittent signal with signal interpolation.

That is, as shown in FIG. 1r, lfield+-fi/delayii! In the if* signal, the scanning line number λ i, +1, i, +! 5, i, +5, river...
Each line image signal of i! after time axis conversion is converted into i! Scanning line number 11+11i1+3 in iigI signal. Earth, +5.
Each line FIJg1 signal j! As such, it is supplied to 118jl after time axis conversion, and a non-delayed image signal! Scanning line number iQ at λ, io+
2, io+4,... each line image signal and l frame delayed image signal! Scan line number i in 3,
, i, +2, 12+4, . i, +2. i, +4.

Interpolation 11i for reconstructing each line image signal of...
i signal! Time axis conversion circuit as B! Supply to I. In addition, these interpolated output images gI Recei! ! and jj, the former is added to a low-pass P-wave filter in the vertical spatial frequency axis direction to extract its low-frequency components, and the latter is also added to a high-pass filter in the vertical spatial frequency axis direction to extract its high-frequency components. By extracting the high and low frequency band components and combining the high and low band components, the scanning line numbers , , i, +2, i, +4 are obtained.

It is also possible to reconstruct each line image signal. According to the configuration shown in Figure 77, a three-dimensional interpolation filter! Each of the input image signals of O has the shape N of a low-speed scanning narrowband image signal as shown in FIG. 3(0), so the signal processing speed of the three-dimensional interpolation filter jO is 1. The signal processing speed of the three-dimensional interpolation filter 7b in the configuration shown in FIG.
The advantage is that the λ configuration of the one-dimensional wave filter is easy.

After performing interpolation F-wave processing as described above, or
The low-speed interlaced narrowband image signal before processing is guided to the time-base conversion circuit si or 7b to undergo time-base conversion, and also performs scanning line interpolation to convert it into a high-speed line-sequential scanning image signal. Convert. An example of the configuration of a time axis conversion circuit that performs such signal conversion is shown in Figure 1 (a).
), and the mode of compression and interpolation of the scanning line is shown in () in the same figure. In the configuration shown in FIG.
! Interpolated output iiiigII signal extracted from! ! and j
An I-line memory 43-/ in which two odd-line image signals j7 and two even-line image signals jl are arranged via a co-pole double-throw switch circuit.

4J-co and 43-J, , 43-4+K each alternately write at low speed, pole double throw switch circuit 60
(b) of the same figure.
As shown in K, high-speed scanning odd line 1lif & signal a
, b, . . . and o, d, . . . are alternately scattered to generate each line image signal a, o, b, d, . Configure.

If the high-speed S*-order scanning wideband III signal, which has been subjected to interpolation P-wave processing and time axis conversion processing as described above, is supplied to a display device t consisting of an ordinary cathode ray tube, etc., the signal will be It is possible to display an image of 11N1 quality that sufficiently prevents image quality deterioration due to the mixing of sampling aliased components that occur due to sampling.

Next, although the signal processing of the present invention is applied to prevent image quality deterioration due to the mixing of sampling aliased components, the configuration shown in Figure 1 is significantly simplified to create a nine-configuration 1iii signal processing device, with the configuration shown in Figure 7. Transmission Confucian composition tatami /
, K, J, and are replaced by a normal interlaced television camera, and the standard television signal obtained from the television camera, i.e., according to the above, the low-speed scanning The interlaced narrowband image signal is converted into a low-speed scanning interlaced narrowband 11i (* used in place of the signal) as shown in FIG. 3(0) in the configuration shown in FIG.
supply to the location. Therefore, with such a simplified configuration, the three-dimensional low-frequency one-wave processing and time axis conversion processing according to the method of the present invention are performed only on the receiving side. In addition, in the signal processing of the present invention method with such a simplified configuration, from the television camera on the transmitting side to the modulation device! Since the image signal supplied to the receiver is the standard television image signal itself, there is an advantage that the image signal can be used separately as the standard television signal.

Furthermore, the above-described ii* signal processing according to the present invention can be applied not only to black-and-white image signals but also to color-iii signals, and basically, each of the primary colors constituting the color image signal It can be applied in parallel to each of the component image signals such as the original signals R, G, and B or the luminance signal Y and each of the color difference signals R-Y and B-Y. However, if three-dimensional one-wave processing to be applied to each of the three component image signals on the receiving side is performed as follows, the number of frame memory devices required in the receiving section 11 can be reduced.

That is, for example, high-speed line sequential scanning color 1
1ii To explain an example of converting an image signal into a high-speed line-sequential scanning wideband color image signal through the process of converting a low-speed scanning interlace color image signal, The sampling frequency points for three-dimensional F-wave processing in the case of conversion to an interlaced image signal are as described above with reference to FIG. Therefore, compared to the luminance signal Y that constitutes the color 1iIi signal, each color difference signal B-Y
.

R-Y (1m! As is well-known, the resolution is extremely low, and the standard color television & (
Since this characteristic of the I signal is actively utilized, the FIJ1fl signal processing according to the present invention also actively utilizes this characteristic. The passband of the two-dimensional low-pass single wave filter used for the interpolation filter 7b on the receiving side is shown surrounded by a dashed line instead of the aforementioned triangular area surrounded by a solid line and shaded. To,
Although a λ-dimensional low-pass V-wave filter having such a passband is constructed using a frame memory and a line memory and is used as a prefilter on the transmitting side using a rectangular region whose width in the vertical intermediate cylinder wavenumber axis direction is halved, the receiving side interpolation filter 7
As for b, only in the vertical spatial frequency axis direction, there is a normal low range on the frequency axis, and in the time axis direction, no low range FIIIS is provided, and a normal color television screen mh
As in the o display, by utilizing the visual low-frequency p-wave effect on the display 1iirt, it is possible to visually suppress the sampling aliasing component in the display image so that it does not cause any disturbance to the color image quality. can. It should be noted that each color difference signal B-Y, which requires a significantly lower degree of resolution,
Regarding R-Y, even if the passband width in the vertical spatial frequency axis direction is significantly stiffer as shown by the dashed line in Figure 1, the image quality of the displayed nine-color image deteriorates substantially. do not. Furthermore, with this configuration, the frame memory for the color signal can be omitted on the receiving side, but the frame memory is also used in the time axis direction for the luminance signal Y, as described above. Since 7'tF wave processing is actually performed, there will be a frame time delay in the luminance signal Y with respect to each color difference signal R-Y, B-Y. It is necessary to add a delay corresponding to the amount of time to each color difference signal R-Y and B-Y in advance on the transmitting side before transmitting them to the receiving side.

Next, wA when the iir signal processing of the present invention is applied to the encoded image signal of the interframe encoding method
An example of the configuration of my signal processing device is shown in FIG. 20(a).

This will be explained with reference to (b). The figure (-) shows the basic configuration of an image signal encoding device using an interframe coding method.
& If the present invention is applied to a signal encoding device, a configuration as shown in the figure) will be obtained. That is, in the image signal encoding device using the interframe encoding method shown in FIG. 20(a),
Input analog image signal! 4I through the adder DD12 and the quantizer t! and its quantizer 4
The quantized gmif* signal obtained from
5 to the frame memory 1744 and read from the frame memory 4t quantization ij
The signal is guided to the adder DD13 and added to the quantized image signal from the doji converter 6j, and the adder DD12
The frame difference image signal is subtracted from the input image signal, and the interframe difference image signal is supplied to the quantizer 6t, thereby forming an interframe encoded image signal and transmitting it to the receiving side. on the other hand,
On the receiving side, the received interframe encoded image signal is supplied to the adder DD14, and the adder output m**
the frame memory 47. In the frame memo I) 47, an image signal of one frame before is written as an initial condition, and the -image signal is read out as a predicted image signal and supplied to the adder DD14, and the input inter-frame image information signal is A new I-frame image signal is formed by adding the seventh I-frame image signal to the frame memory 47, so that the output frame image g1 signal 4F is sequentially extracted from the adder ADD14. It is. On the other hand, in the interframe image signal encoding device to which the image signal processing method of the present invention shown in Figure Φ) is applied, r! 1lF
jA(a) As an input image signal evaluation on the transmitting side in the conventional device with the basic configuration shown in FIG. A low-speed scanning interlaced narrowband image arrangement as shown in Fig. 3 (0) is used, and on the receiving side, the frame memory t7 for extracting the weave image signal is used only for forming the predicted signal. 7th without using
It is also used as a frame memory for the three-dimensional Fukuma filter 7b in the illustrated configuration. In other words, line sequential scanning method! For example, in the case of converting an image signal into an interlaced image signal, the frame memory 7 is divided into two to form one field memory, and these field memories are configured as shown in FIG. In the receiving side signal processing device according to the present invention, the receiving side device in the image signal 7-ray human encoding device having the basic configuration shown in FIG. By having exactly the same configuration as the circuit device on the side of the signal processing device of the present invention method 1i shown in Fig. 77, the present invention - image signal By performing three-dimensional interpolation filter processing using the processing method, it is possible to prevent deterioration in the quality of the reproduced image due to the mixing of sampling aliased components caused by quantization during encoding.

First, the present invention 11i (*The signal processing method is based on frame-related
In the same manner as described above for the image signal encoding device, it can be applied to a receiving side circuit device in so-called still image broadcasting, and the same effects as described above can be obtained.
In a transmission system for still image broadcasting, a frame memory is provided on the receiving side, and a desired still image signal among nine different still image signals transmitted one frame at a time is stored in the frame memory and repeatedly reproduced and displayed. Therefore, even in still image broadcasting, the frame memory provided on the receiving side is divided into two field memories, and these field memories are shown in FIG. If similar signal processing is performed using the L field memory box, lit, in the processing device according to the configuration shown in Figure 77, high-quality images can be reproduced and displayed by performing IFAKJ-dimensional interpolation processing. .

Next, we will discuss an example in which the image signal processing of the present invention is applied to O contour compensation to improve the sharpness of a television image, and contour compensation is performed for three-dimensional wave number components including the time axis. explain.

Conventionally, in order to improve the sharpness of a television set, an adder module DD is used to combine the input picture signal and the l-line delayed image signal, which is cascade-connected to the 9 I-line delay lines 7 and 72', to form the input picture flaw signal.
In addition to the 1B and j dB attenuation 973, an average image signal is formed, and the average jIgI signal and the l-line delayed image signal are added together.
By adding the obtained difference-image signal in addition to the l-line delayed image signal by an adder DD 20, contour compensation in the vertical spatial frequency axis direction is performed and a blank output image signal 7j is obtained. Therefore, the spatial frequency characteristics of such a contour compensation circuit are as shown in FIG. Basically, by using an l-line delay line, a vertical spatial frequency characteristic related to the number of scanning lines NK is obtained as shown in the figure.Also, contour compensation in the horizontal spatial frequency axis direction is obtained. , /line delay lines 7λ, 72' in the configuration shown.
Instead, a delay line having a delay amount of several hundreds of nanoseconds is used to make the horizontal spatial frequency characteristic match the above-mentioned - direct space function wave number characteristic. As shown in Figure 1, the vertical spatial frequency characteristic is such that the center frequency of the boost region for contour compensation is M7. It is located at In addition, in FIG. 22, the spatial frequency of the television screen il* is normalized and expressed by the height of one surface, and in an image with the number of scanning lines N,
It is said to be capable of transmitting and displaying up to 172 spatial frequencies. Therefore, the center frequency N7 of the rounded spatial frequency characteristic boost region for sharpness improvement obtained by the conventional contour compensation circuit by Sei Kusakaki shown in FIG. had the disadvantage that it could not be transmitted or displayed.

In order to eliminate such conventional drawbacks and increase the center frequency of the vertical spatial frequency characteristic boost region, it is possible to use a 7-field delay instead of the l-line delay line. When contour compensation is performed in the vertical spatial frequency axis direction using the above-described configuration, extremely strong flicker disturbance occurs in the details of the image in a standard television image that performs line interlacing, and on the contrary, the display image quality is significantly degraded. It turns out. In addition, in conventional television camera,
Since the accumulation effect of the camera tube is used to increase the sensitivity, the i! Since severe blurring occurs in iif*, this significantly deteriorates the display image quality.

In order to perform the above-mentioned contour compensation related to the contour compensation of the television image, a circuit device having exactly the same configuration as that shown in FIG. 7 is used, as shown in FIG. 23.

To briefly explain the configuration shown in Fig. 1, the Sm-order scanning direction signal with the number of scanning lines H from the camera I is supplied to a three-dimensional low-pass F wave generator λ to limit the signal spectrum distribution area. Then, the signal is supplied to a time axis conversion circuit, where it is converted into a Coni/line interlaced system III image signal, and further supplied to a modulation device JK, where it is converted into a carrier signal. However, 3
The required passband of the low-pass P-wave device is 1!, as shown in the first reference. As detailed in , it is a triangular area with diagonal lines. That is, in the frequency domain distribution shown in the second figure, the signal spectrum distribution in the frequency domain surrounding the sampling frequency point with a dotted line is the signal spectrum distribution of the frequency domain where the sampling frequency point is surrounded by a dotted line. It occurs in the ninth place,
In addition to the camera output image flaw signal, each point 0. D
, , , etc., and these unnecessary signal components mix into the original signal spectrum distribution region including locus III origin B as so-called aliasing distortion components. In order to prevent this, a three-dimensional low frequency FULLI device is provided. Note that when converting to a 2;7 line interlaced picture signal, there are essentially 2
It acts as a low-dimensional #F wave device as described above.゛' Finally, on the receiving side, the J
:l The line interlace system jI signal is reconverted into a line sequential scanning system image information signal by the time axis conversion circuit 7a, and the converted output iiiI signal includes each point in the frequency domain distribution shown in the figure, Sideband components in the frequency region surrounding 0°D are included, and if those sideband components adjacent to the frequency region surrounding point B are displayed, the reproduced image quality will be significantly degraded. A 'dimensional low-pass single wave 41 is provided as an interpolation filter 7b to sufficiently suppress such unnecessary signal components. In general, sideband components in the frequency domain surrounding point 0 in the frequency domain distribution shown in the second figure are naturally suppressed due to the 10% nature of the display device and the decrease in visual MTF characteristics in the vertical spatial frequency axis direction. Also, the @band wave component in the frequency domain surrounding the point Dt is the visual MTff% in the time axis direction.
Since the entire transmission system is designed in such a way that it is naturally suppressed by a decrease in the frequency, it usually does not significantly impede the reproduced image quality. do.

Next, the case where the contour compensation according to the method of the present invention in the above-mentioned aspect is applied to a television image information transmission system will be explained. In addition to the contour compensation, contour compensation is also performed in the time axis direction, and for the j:l line interlaced image signal, the signal spectrum distribution limited region shown in the figure is not made to have a simple low-pass characteristic, but a pass-through Second contour compensation is performed by boosting the vicinity of the upper limit frequency of the band. The mode of such contour compensation is shown in Fig. Jj (J
They are shown in sequence in L), (b), and (0). In other words, this is a book that expands the one frequency region near the diagonal line in the above-mentioned triangular required passband shown in FIG.
indicates a boost characteristic in the -direction spatial frequency axis direction, and
(0) in the same figure shows the boost characteristic in the time axis direction.

Therefore, the three-dimensional band limit Fall shown in FIG.
! Although the boosting mode of P [it
needs to be configured. Therefore, it is preferable that the three-dimensional band-limited single-wave characteristic shown in FIG. In order to realize the combination of p-wave characteristics, a circuit configuration as shown in FIG. 27 is used. That is, in the configuration shown in FIG. 27, 7r
is a band-limited P-wave device in the time axis direction, and its passband characteristic is the amplitude/frequency characteristic-II (a
), and boosts the region (~) in the passband characteristic shown in Figure 14. Also, 7v in the configuration shown in Figure 27 is a band control @ F-wave device in the direction of the open wave number axis in the vertical space. Therefore, the passband characteristics are as shown in the amplitude/frequency characteristics -llΦ) shown in the first figure, and the first boosts the area φ) in the passband characteristics shown in the figure. Furthermore, 10 in the configuration shown in FIG. 27 is a band control device in the time axis direction, and its pass band characteristics are the amplitude/frequency range shown in FIG. ＃The characteristic curve (a) looks like this,
The region (0) in the passband characteristic shown in FIG. 24 is boosted. Further, r/ in the configuration shown in FIG. 27 is vertical 9
It is an aPIN device with band control in the direction of the same wavenumber axis, and its passband characteristics are as shown in the amplitude/frequency characteristics shown in Figure 8 (d). ) to boost. Note that all of these band-limiting@-wave devices can be easily realized, for example, by a so-called transversal filter having a configuration similar to that shown in the first diagram. 71°to can use a memory element for the I field of a progressive scanning system image signal as a unit delay element of a transversal filter, and also - band control in the vertical spatial frequency axis direction @F wave device lid, 1/, a memory element corresponding to l life of the sequential scanning image signal can be used as a unit delay element of the transversal filter.

In addition, the required passband characteristics shown in Figure 11 (&JK) are
It is preferable that K, which is approximated more precisely than the approximate characteristic shown in Fig. 4, be a combination of approximated passband characteristics as shown in Fig. Jσ, for example.

That is, as shown by diagonal lines in Fig. 3Q, the dead area is boosted by approximating the staircase shape of minute steps, and as in the case 9 shown in Fig. 27, even more types of amplitude @ frequency characteristics are obtained. It is constructed by combining a large number of band-limiting F-wave devices in the time axis direction and the vertical spatial frequency axis direction.

Note that contour compensation in the horizontal spatial frequency axis direction is performed in exactly the same manner as the conventional method described above, and is performed within several hundred nanoseconds.
! A contour compensation circuit in the horizontal spatial frequency axis direction is constructed using a delay line having a delay amount of 100.degree. use

9, the boost frequency region in this case is preferably a frequency region that corresponds to region (b) in the passband characteristic shown in FIG. 8 among the boost frequency regions in the vertical spatial frequency axis direction.

In order to simplify the configuration of the contour compensation circuit according to the present invention as described above, the first region (
It is also possible to omit both or one of the band-limiting P-wave devices in the time axis direction corresponding to a) and (0). That is, in the configuration shown in No. 27IgI, the band limiting @F wave device 7t in the time axis direction corresponding to the region (a) in the passband characteristic shown in FIG. 4 is relatively superimposed and is difficult to omit; Although the increase in the number of circuit elements due to the provision of this P-wave device 71 is relatively small, the band limit @F-wave device in the time axis direction corresponding to the region (0) in the passband characteristic shown in Fig. 21 is 1
If 0 is omitted, the number of circuit elements can be significantly reduced.

First, in the above explanation, we will discuss contour compensation applying jl (I signal processing) according to the method of the present invention only as an example of converting a line-sequential scanning system jl signal to a line interlaced system ms signal. However, the contour compensation according to the method of the present invention not only depends on the interlace ratio J:/;
When converting to a line interlaced image signal with an even larger interlace ratio, or to a dot interlaced image signal*, the same method as described above can be applied. In this way, the required passband of the three-dimensional band control@-wave device is set, and the area near the upper limit of the required passband is shown in Figure 21 (
Contour compensation can be performed by boosting in the same manner as in item 9 shown in a).

Therefore, 1 itr to which the image signal processing method of the present invention is applied
*As with contour compensation for conventional standard television images, contour compensation only needs to be performed on the transmitting side, and there is no need to perform it individually on the receiving side, but if it is also performed on the receiving side, it will be even more effective Contour compensation can be performed in a suitable manner, and in that case, a required boost characteristic is imparted to the passband of the three-dimensional band-limiting VkP waveform device constituting the interpolation filter 7b on the receiving side in the configuration shown in Fig. Let's do it. Note that when three-dimensional contour compensation is performed by providing a three-dimensional interpolation filter on the receiving side and boosting its passband in this way, the three-dimensional contour compensation on the transmitting side may be omitted. Therefore, even when using a one-line interlaced imaging camera on the transmitting side as in the past, the image signal processing of the present invention can be applied to contour compensation to achieve sufficient effects. I can do it.

As is clear from the above description, 1. According to the present invention, on the transmitting side of the television transmission system, the original signal spectral distribution region of the three-dimensional sampling aliasing distortion component that occurs due to the scanning of the weave camera Therefore, it is possible to sufficiently eliminate the deterioration in reproduction display quality due to the mixing of sampling aliasing distortion components, and it is possible to perform high-quality image transmission. In addition, on the receiving side, display uniformity is achieved by removing three-dimensional sampling frequency points and their sideband components included in the received signal or generated by signal processing on the receiving side using a three-dimensional interpolation filter. , it is possible to display an image with excellent image quality consisting only of the original baseband frequency component, and significantly reduces the image quality deterioration caused by the contamination of the 93-dimensional sampling horse wave number component and its sideband wave components, which were previously difficult to sufficiently remove. It is possible to improve the image quality and display high-quality images. If the removal of unnecessary signal components by such a three-dimensional wave device is applied to both the transmitting side and the receiving side of the TV 17 image transmission system, the most desirable image transmission system can be constructed in principle.

In addition, if a three-dimensional interpolation filter with the minimum performance necessary to sufficiently suppress unnecessary signal components is used on the receiving side, the image quality of the reproduced display can be improved, and the three-dimensional interpolation filter on the receiving side can be improved. Since the band limit (wave) can be simplified and made small, the cost of the receiving circuit device can be reduced.

Furthermore, on the receiving side, 1ji is calculated using a three-dimensional interpolation filter.
If f & signal processing is performed and the time axis conversion of the iii signal is performed in the ninth stage, the required speed of image signal processing by the three-dimensional interpolation filter can be lowered, and it is possible to reduce the friction of the receiving side circuit device. It can be made into a product.

In addition, the transmitting side does not perform any special signal processing such as 1-j three-dimensional band limiting, subsampling, or time axis conversion, and transmits the normal standard television broadcast signal as is, and the receiving side By performing the three-dimensional interpolation filter processing and time axis conversion as described above, the image signal processing of the present invention can be applied to a normal TV set.
Can be applied to signal transmission.

Three-dimensional band limiting, subsampling, and time axis conversion are applied to the image output image signal of the line sequential scanning method.The image signal originally has almost the same signal form as the interlaced image signal. Therefore, the image signal of the process m* according to the present invention can be directly transmitted to the ordinary standard television system myijI.
It has the advantage of being able to be used interchangeably with mine. That is, the sending fI@output 1iii signal processed by the configuration shown in the first diagram is as follows:
If it is converted to the line interlaced format, it can be used as a standard television signal as it is, and the receiving side can perform one-dimensional interpolation filter processing and time axis conversion using the configuration shown in Figure 7. Without,
Even when reproduced and displayed in the conventional Doosho 11 format, it is possible to display an image with much better image quality than the conventional one, with less deterioration in image quality due to the mixing of sampling aliasing distortion components.

In addition, when applying the image signal processing of the present invention to color image signals, the frame memory required for three-dimensional V-wave processing on the receiving side can be simplified for color signals with narrow frequency bands. The capacity of the frame memory is not so large as can be inferred from the required capacity of the frame memory for processing black-and-white signals, and the receiving side circuit device can be made simpler and cheaper.

In addition, when applying the image signal processing of the present invention to interframe coding of 1ii (# signal), three-dimensional band limitation is applied to the signal form before interframe coding processing is applied to the image signal on the transmitting side. Since it is applied in accordance with On the other hand, since the frame memory for forming the predicted frame 1iii customary code can be used as the frame memory for one-dimensional filtrate processing according to the method of the present invention, the required number of frame memories can be reduced and the method according to the present invention can be used. By combined use of image signal processing, the decoded output signal with significantly improved image quality can be reproduced and displayed using a relatively low power one-way device.

(9) Even when applying the Ill signal processing of the present invention method to still image broadcasting, the present invention method may be used to provide a 7-ray 5 memory for repeatedly reproducing and displaying a desired still image signal frame by frame in the receiver 111. Since it can be used as a frame memory Kll for three-dimensional reactor liquid processing, still images with significantly improved image quality can be reproduced by the image signal processing method of the present invention without significantly increasing the manufacturing cost of the receiving side circuit device. This can have a significant effect on improving frame memory usage efficiency and popularizing still image broadcasting services.

Furthermore, when applying the signal processing of the present invention to contour compensation of an image, conventionally, the frequency domain in which boosting is applied for contour compensation in the direction of the vertical spatial frequency axis is
However, by using the image signal processing method of the present invention in conjunction with the image signal processing method of the present invention, it is possible to improve the vertical spatial frequency for images with little temporal change. The boost region for contour compensation in the frequency axis direction can be increased to the frequency region near λ,
In addition, for images with large movements, 4.N7.
Since the boost frequency region in the vicinity can be secured, the image quality improvement effect by contour compensation can be significantly increased compared to the conventional method, and furthermore, by removing sampling aliasing distortion components, it is possible to reproduce and display extremely high-quality images. I can do it.

Furthermore, if the iij* signal processing of the present invention is used in conjunction with image contour compensation, contour compensation will also be applied to the frequency characteristics in the time axis direction, so if the subject moves, the contour portion will be forced. Therefore, it is possible to significantly improve image quality deterioration due to blurring of moving images, which conventionally occurred in such cases due to the accumulation effect of the camera tube.

It should be noted that since it is only necessary to apply the image contour compensation using the IIR* signal processing of the present invention in combination with the image contour compensation, there is no need to individually perform contour compensation on the receiving side, making the entire system economical. Although it can be configured as follows,
On the receiving side, when using a three-dimensional band control @wave device, contour compensation using the above-mentioned ILI signal processing according to the present invention method can be performed without particularly increasing the price of the receiving circuit device. , can be performed individually on the receiving side, so the transmitting side can remain as usual, and a high-quality image can be reproduced and displayed by image signal processing only on the receiving side.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the basic configuration of an image signal processing device according to the present invention, FIG. 2 is a diagram showing the dynamic principle of the same signal processing, and FIG. 3 is a block diagram showing the various parts shown in FIG. 1. A waveform diagram showing an example of a signal waveform, the second figure is a characteristic curve diagram showing an example of the frequency characteristics of each part of the signal, and FIG. 1 is an example of the detailed configuration of the time axis conversion (b) path in the configuration shown in the first figure. FIG. 6 is a diagram showing the aspect of the three-dimensional configuration of 1ljif* to which the image signal processing according to the present invention is applied, and FIG. 7 is the frequency component distribution in the three-dimensional configuration of the image. Figure 2 is a diagram showing an example of the required passband of a three-dimensional low frequency door door in the configuration shown in Figure 1, and Figure 2 (a) and Φ) are diagrams showing the required passband of the three-dimensional low frequency A diagram and a block diagram showing an example of an actual passband and an example of a circuit configuration of a PM device, the i-th
o Figures (a) and Φ) 4 Diagrams and block diagrams respectively showing other examples of the actual passband and circuit configuration of the three-dimensional low-pass single wave filter, Figure 1/ is the same. 1 that 3-dimensional low frequency single wave device! The first λ diagram is a block diagram showing an example of the detailed configuration of an actual low-band F wave device to be viewed, and the first λ diagram is a diagram showing another example of the required passband of the three-dimensional low-band F[G.
Figure 1J is a diagram showing still another example of the required passband of the three-dimensional low-pass F wave filter, and Figure 7 is a diagram showing the λ-dimensional configuration of an image to which image signal processing according to the present invention is applied. A microcosm showing the aspect, No. 1! The figure also shows other aspects of the frequency component distribution in the three-dimensional low-frequency ν-wave device. Diagrams and block diagrams showing still another example of the required passband and still another example of the circuit configuration, respectively; FIG. 77 shows an example of the receiving side configuration of the image signal processing device according to the method of the present invention. A block diagram, FIG. 17 is a signal waveform diagram showing an aspect of image signal processing according to the configuration shown in FIG. 17, and FIG. Block diagrams and signal waveform diagrams showing aspects of signal processing, respectively;
Fig. 10 (~ and Φ) is a block diagram showing an example of the configuration of a 1lif & signal encoding device applying image signal processing according to the present invention, and Fig. FIG. 23 is a characteristic curve diagram showing the compensation characteristics; FIG. 23 is a block diagram showing a configuration example of an ms contour compensation circuit to which image signal processing according to the present invention is applied; and FIG. 23(a) and Φ) and (0) are diagrams and characteristics respectively showing an example of the required passband and boost characteristics of the three-dimensional low-frequency fall used in the ii*contour compensation circuit. The curve diagram, FIG. 26, is a reduced scale diagram showing an example of the actual passband of the three-dimensional low-frequency wave device, and FIG.
Fig. 21 is a block diagram showing an example of the detailed configuration of No. 5, Fig. 21 is a characteristic curve diagram showing another example of the boost characteristics of the J-dimensional low-area reactor, and Fig. 21 is a characteristic curve diagram showing another example of the boost characteristics of the FIG. 30 is a characteristic curve diagram showing still another example, and FIG. 30 is a diagram showing another example of the actual passband of the three-dimensional low-range furnace liquid vessel. l: Photographer camera, ko: 3-dimensional prefilter, 3
...Subsample circuit, 7a...Time axis conversion circuit,! ...Modulation device, t...Demodulation device, 7b...
・Three-dimensional interpolation filter, l...display device, l...input image signal, 10, /7...low pass F
Wave converter, no...Analog-to-digital converter, lco,
/j. /F, KO...changeover switch, JJ, /reference switch
...l line delay line, ll...digital-to-analog converter, ll...output image signal, λc...input clock signal, here...1 frequency divider, koJJ...input image signal, Co reference...Low pass p-wave filter in the time axis direction, λj...Low pass filter in the vertical frequency axis direction, -1...Output iii* signal, Core...Input image signal, JZ, , 2... Time axis direction low pass P-wave device, 10, Jl... Vertical frequency axis direction low pass - wave device, JJ... Output image signal, JJ... Input image signal, 3 Reference...l frame or l line delay line, 3! ...Weighting coefficient unit, 3z/output image signal, 77, Sanco...Input III signal, Jl...Low pass in the time axis direction - wave unit, Jri...Low pass in the vertical frequency axis direction Passage paS%UO...Horizontal frequency axial low pass p
Wave equipment, dream! ,case! ...Output image signal, see 3. Da≠...
・Three-dimensional low-frequency PTL device, reference t, j...input iii my signal, Hiro 7. ≠l...
l field filter, Jl... time axis conversion circuit, jJ
...I-Fre-Taigo, jJ, #...Interpolation output signal, j7, ! I...Interpolation output image information signal,! ri, 4
0, 4/... changeover switch, λ... output jl
signal, 4J-/~4J-1 line memory, 4 signal... input 1lii signal, 47... quantizer,
44, A7...Frame memory, 41...Output image signal, 7)...Input 1i1 (# signal, 72.7λ
'...l line delay line, 73...attenuator, 7hiro...
- Variable attenuator, 7j... Output image signal, 77... Input image signal, 7r, to... Time axis direction low pass P wave device, 79, r/... Vertical frequency axis direction low frequency Passing P wave device, lco...output image signal, mu DD1 to mu DD22... adder. Patent applicant Japan Broadcasting Corporation Figure 1 Figure 2 Dire Figure 3 Figure 1 Snout Figure 5 Figure 6 Figure 7 Horizontal dark soft glaze Figure 8 Hole Figure 9 Figure 12 Tag Figure 14 Figure 15 4ν Fig. 16 (a) v Cb) l,ri5σ (C) 2 Fig. 17 4σ Fig. 18 Fig. 19 Fig. 21 JAJJh=F) Normal! t-t*ymhatl,? l [Fig. 23, Fig. 24, 4ν, Fig. 25, child V fL4 conversion to vertical sky 1'l/1 frequency axis no darkness I! I! Flr.
l-Axis Procedural Amendment Written April 8, 1982 1. Display of the case 1982 Patent Application No. 204189 2. Name of the invention Image signal processing method 3. Person making the amendment Relationship to the case Patent applicant ( 435) Japan Broadcasting Corporation 5゜6・Object of amendment Detailed explanation column of the invention in the specification, drawing 7, contents of amendment (as attached)
r,! , UP of specification No. 66) ii-th line to 14th line
111 when converting f, therefore, `` is #11I
There will be a lot. Of the two drawings, Figures 8 and 9 (a) are corrected as shown in the attached sheet.

Claims (1)

[Claims] 1. Based on the sampling of the image signal in a multi-dimensional coordinate consisting of a coordinate plane formed by at least some of the horizontal frequency axis, vertical frequency axis and time axis regarding the image signal. At least one plurality of at least one sampling frequency point whose desired passband is a region on the origin side of the multidimensional coordinates of a line connecting at least one sampling frequency point that occurs and points that are approximately equidistant from the origin of the multidimensional coordinates. Dimensional low frequency band 41a: An image signal processing method using a mackerel to suppress high frequency unnecessary signal components on the coordinate plane formed by at least the vertical frequency axis and the time axis, which are generated by sampling the image signal. . By making the transfer function of the high frequency section of the passband larger than the transfer function of the low frequency section on the coordinate plane formed by the vertical frequency axis and the time axis, the signal component of the high frequency section can be increased. An image signal processing method that uses % mold. 3. In the image signal processing method according to claim 1 or 2, the multidimensional low-pass P-wave device is provided on both the transmitting side and the receiving side of the image signal transmission system, respectively. An image signal processing method characterized by: 4. In the i & i/o signal processing method according to claim 1/2, the multi-dimensional low-pass P-wave device is provided only on the transmitting side of the Aii/i signal transmission system. An image signal processing method characterized by: 5. The image signal processing method according to claim 1 or 2, characterized in that the multi-dimensional low-pass P-wave device is provided only on the receiving side of the 1lif & signal transmission system. Image signal processing method.