JP2000244877A - Image signal motion compensation double speed conversion circuit and television device - Google Patents

Abstract

(57) [Summary] [PROBLEMS] To provide a double speed conversion method and circuit of an image signal with high image quality and high resolution. SOLUTION: A still interpolation generation unit 4 generates an interpolation signal SIP suitable for a still image by inter-field processing. The VT interpolation generator 5 generates an interpolation signal VTIP suitable for a moving image by vertical / temporal filter processing. The MC interpolation generation unit 6 generates an interpolation signal MCIP suitable for slow motion and vertical pan motion by GST (General Sampling Theory) processing using the motion vector PV. The setting unit 7 generates a multiplex control signal CT based on the magnitude of the motion vector PV and the prediction error MVE. Selector 8
Generates an interpolated scanning line signal IP by a weighted addition process of a mixing ratio determined by the multiplex control signal CT. The speed doubling unit 10 includes a signal S3 whose time delay is adjusted by the delay unit 9 and a signal of the interpolation scanning line.
It performs 1/2 compression of the time axis of the IP and time-series multiplexing processing, and outputs a sequentially scanned image signal S4.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a double-speed scanning conversion method and circuit for converting an interlaced scanning image signal into a progressive scanning image signal by a scanning line interpolation process, and more particularly to a resolution reduction with a relatively slow movement. The present invention relates to an image signal motion compensation double-speed conversion method and circuit suitable for realizing high-quality jump-to-sequential scan conversion with little interference or flicker.

[0002]

2. Description of the Related Art In recent years, progressive scanning has been adopted as a scanning mode in multimedia TVs, which are PC / TV integrated information home appliances, and display devices such as PDPs (plasma displays) and LCDs (liquid crystal displays). However, most of the image signals are captured in the form of interlaced scanning. For this reason, a function of double speed conversion for performing interlaced to sequential scan conversion with high image quality is required.

As a typical one of the interlaced to sequential scan conversion, a method using a motion adaptive process is known. this is,
A mixing ratio between an interpolation signal suitable for a still image generated by the inter-field interpolation processing and an interpolation signal suitable for a moving image generated by the interpolation processing in the field is changed according to the motion of the image, A signal of a scanning line that is skipped by the interlaced scanning is generated. However, this method has a problem that the resolution characteristics of the moving image are poor, and the vertical resolution is degraded by certain kinds of motions, and unpleasant flicker interference occurs.

[0004]

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned problems, and has been made in consideration of the above-described problems, and has been made in view of the above-mentioned problems. It is an object of the present invention to provide a signal double speed conversion method and circuit.

[0005]

Means for Solving the Problems In order to achieve the above object, the present invention employs the following technical means.

The interpolation signal includes a first interpolation signal by inter-field signal processing suitable for a still image, a second interpolation signal by motion correction signal processing suitable for a slowly moving image, and a moving image. Vertical / temporal filter (VT)
And a third interpolation signal by signal processing in the field. Note that the slow motion refers to, for example, a motion in which two lines or two per field are smaller than the original.

Among them, the first and third interpolation signals are signals widely used in the prior art. on the other hand,
The interpolation signal by the second motion compensation is GST (General Sample).
ingTheory).

GST is a generalized theory of the well-known sampling theorem, and shows that an original signal can be completely decoded with two types of sampled signal sequences having different phases. If this theory is applied to a signal sequence of two fields of interlaced scanning having different sampling phases in the vertical / time domain, it is possible to decode a signal of a scanning line which has been skipped in interlaced scanning. An outline of this principle will be described with reference to FIG. The signal sequence of the scanning line in the current field L1 of the interlaced scanning signal (the signal of the scanning line indicated by the symbol 中 in the figure) is..., L1 (2kh-3h), L1 (2kh
−h), L1 (2kh + h),..., Where the signal sequence of the scanning line of the previous field L2 is moved to the position of the current field by the motion vector y (the scanning line indicated by the diamond in the figure). Line signal)
……, L2 (2kh−4h + y), L2 (2kh−2h + y),
L2 (2kh + y), ... According to the GST, the signal L1 (2kh) of the interpolated scanning line (the signal of the scanning line indicated by a black circle in the figure) is calculated by the following Equation 1 for the two sampled signal sequences having different phases. Can be decrypted.

L1 (2kh) = ΣV1 (j) · L1 (2 (k + j) h−h) + ΣV2 (j) · L2 (2 (k−q + j) h) Equation 1 Here, the motion vector y = 2h ( q + r), 2h is the distance between scanning lines in interlaced scanning, q is an integer, and 0 <r <1.
Σ is the sum of j = −∞ to ∞.

Note that the coefficients V1 (j) and V2 (j) in Equation 1 are given by Equations 2 and 3, respectively.

V1 (j) = (− 1) ** j · sinc {π (j−0.5)} · sin (πr) / cos (πr) Equation 2 V2 (j) = (− 1) ** j · sinc {π (j + r)} / cos (πr) Expression 3 In Expressions 2 and 3, sinc {x} represents sin (x) / x.

Therefore, in the generation of the first and third interpolation signals of the prior art, the degradation of the resolution and the occurrence of flicker disturbance cannot be avoided. Using the second interpolation signal generated by the correction processing, it is possible to realize high-quality, high-resolution jumping to sequential double-speed conversion, which is difficult to achieve with the conventional technology.

In order to detect a motion vector (accuracy of 2 rh in the vertical direction per field) required for the motion correction processing, in the present invention, a motion vector is detected in a plurality of frame periods. During the two-frame period,
H / 4 per field, h per field during 4 frame periods
Enables vertical motion detection with / 8 accuracy.

Further, in order to suppress the image quality deterioration peculiar to the motion correction process caused by the erroneous detection of a motion vector (detection of a motion different from the original motion), the present invention uses The accuracy of the motion vector is evaluated, and when the accuracy is low, signal processing for restricting the use of the interpolation signal in the motion correction process is performed.

By the technical means described above, it is possible to realize a jumping-to-sequential double-speed conversion method and circuit having high image quality and high resolution characteristics which are difficult to achieve with the prior art.

[0016]

DESCRIPTION OF THE PREFERRED EMBODIMENTS A first embodiment of the present invention will be described.
This will be described with reference to FIGS.

FIG. 1 is a block diagram showing an example of this block structure. A frame delay section 1, a motion vector detection section 2, an accuracy verification section 3, a still interpolation generation section 4, a VT interpolation generation section 5, an MC interpolation generation section 6,
It comprises a setting unit 7, a selection unit 8, a delay unit 9, a speed-doubling unit 10, and a control unit 11.

One of the interlaced scanning image signals S1 (luminance signal and color difference signal) is input to the control unit 11 to generate signals necessary for signal processing operation. To supply.

On the other hand, the frame delay section 1 generates a signal S2 obtained by delaying the input signal by a predetermined frame period.
The period of the frame to be delayed depends on the detection accuracy of the vertical motion used in the motion compensation processing. For example, h / 4 per field (interval between scanning lines in interlaced scanning is 2
h, the interval between scanning lines is equivalent to h in the case of sequential scanning). If the accuracy is required, set to 2 frame periods, and if the accuracy of h / 8 is required, set to 4 frame periods. .

The motion vector detecting section 2 detects a motion vector for each block by block matching processing.
In the mini-block division search process, a motion vector PV in pixel units required for the motion correction process is generated. Further, the accuracy verification unit 3 calculates and outputs a prediction error MVE obtained from the motion vector PV.

The still interpolation generator 4 generates an interpolation signal SIP suitable for a still image by signal processing between fields. Further, the VT interpolation generation unit 5 generates an interpolation signal VTIP suitable for a moving image using a VT filter in a time / vertical region. On the other hand, M
The C-interpolation generating unit 6 generates an interpolation signal MCIP suitable for a slowly moving image by the arithmetic processing of Expressions 1 to 3 according to GST based on the motion vector PV.

The setting unit 7 sets a multiplex control signal C for designating a coefficient value of coefficient weighting based on the motion vector PV and its prediction error MVE.
Generate T. Then, the selection unit 8 outputs the multiplex control signal C
Weighting and adding the coefficient value determined by T, the signal IP of the interpolation scanning line
Generate

The speed doubling unit 10 performs a 1/2 compression process and a time series rearrangement process on the time axis of the signal S3 in which the time delay accompanying the signal processing is adjusted by the delay unit 9 and the signal IP of the interpolation scanning line. An image signal S4 converted into a progressive scan is generated.

The configuration and operation of each block will be described below.

FIG. 2 is a block diagram showing an example of the configuration of the motion vector detecting section and the accuracy verifying section. The block motion vector detecting section 12, the singular vector correcting section 13, the memory section 14, the mini-block division searching section 15, and the prediction error calculating section. 16.

The block motion vector detecting section 12 outputs a signal
Using the luminance signal components of S1 and S2, or the luminance signal component and the color difference signal component, first, the detected motion vectors in the blocks around the target block (block size is, for example, 8 horizontal pixels × 8 vertical lines) are referred to as reference vectors. RMV is set as the representative vector with the smallest prediction error. Note that the prediction error ER is a signal
The signal of the current frame composed of S1 and the signal several frames before composed of signal S2 are referred to as a reference vector RMV (horizontal components RMVx,
It is calculated by the sum of the absolute values of the difference components from the signal of the point whose position has been moved by the vertical component RMVy). If this is represented by a mathematical expression, it becomes Expression 4.

ER = Σ | S1 (x, y) −S2 (x + RMVx, y + RMVy) | (4) When the prediction error of the representative vector is smaller than the set value, this is set as the motion vector BV of the target block. On the other hand, when the value is equal to or larger than the set value, a predetermined search area is searched for by the block matching process starting from the representative vector, and the one having the smallest prediction error is set as the motion vector BV of the target block.

The singular vector correction unit 13 examines the correlation between the motion vector of the target block and the motion vector of the peripheral block, and performs processing of correcting a motion vector with a small correlation to a motion vector with a high correlation. That is, a motion vector of the target block different from any of the motion vectors of the peripheral blocks is detected as a singular vector, and the singular vector is replaced with a motion vector of a peripheral block having the smallest prediction error. Then, a highly accurate motion vector BV1 is output. One of them is stored in the memory unit 14 and used as a reference vector RMV.

The mini-block division search section 15 uses a motion vector of the current block and a peripheral block to generate a mini-block (for example, 2 pixels × 2 lines) obtained by subdividing the block in the horizontal and vertical directions. Generates a motion vector PV of the pixel of. That is, for a block in which the prediction error due to the motion vector of the current block is less than the set value, the motion vector is used to generate the motion vectors of the pixels of all the mini-blocks in the block. On the other hand, for a block equal to or larger than the set value, for each mini-block, the motion vector of the current block and the peripheral block having the smallest prediction error is detected, and the motion vector of the pixel in the mini-block is generated.

On the other hand, in the accuracy verification section, the prediction error calculation section 1
6. For each pixel, the prediction error obtained by the motion vector PV
MVE is calculated by the following equation 5. PVx, PVy in the formula
Are the horizontal and vertical components of the motion vector PV.

MVE = | S1 (x, y) −S2 (x + PVx, y + PVy) | Equation 5 Next, the stationary interpolation generator and the VT interpolation generator are shown in FIG.
Will be described.

FIG. 3A shows an example of the impulse response of stationary interpolation and VT interpolation in the vertical / time domain. In the figure, white circles indicate scanning line signals sent by interlaced scanning, and black circles indicate interpolation scanning line signals. In the static interpolation, an interpolated signal is generated by a signal of a scanning line of the previous field at the same position as the interpolated scanning line (a scanning line having a coefficient of 1 in the figure) in inter-field processing. In VT interpolation, the vertical low frequency component of the interpolation signal is obtained by multiplying the signal of the scanning line of the current field by 1/16, 7/16, 7/16, 1/16, and the vertical high frequency / time low frequency The component is generated by multiplying the signal of the scanning line of the previous field by coefficients -4/16, 8/16 and -4/16.

FIG. 3B shows an example of this configuration, in which the 1H delay unit 1
7, 260H delay unit 18, coefficient weighting unit 19, addition unit 20
It consists of.

The signal S1 is supplied to the 1H delay unit 17 for one scanning line period,
The coefficient weighting unit 19 applies coefficient values 1/16, 7/16, 7/1 to the signal sequence delayed by 260 scanning line periods by the 260H delay unit 18.
6, 1/16, -4/16, 8/16, -4/16 are weighted and added by the adder 20, and an interpolation signal VTIP by VT filter processing is obtained from this output. Further, an interpolation signal SIP is obtained from the output of the 1H delay unit 17 by the inter-field processing. Although not explicitly shown in the drawings, the same configuration is required for each of the luminance signal and the two color-difference signals (even one signal obtained by time-division multiplexing is possible).

Next, the MC interpolation generator will be described with reference to FIGS.

FIG. 4 shows an example of this configuration, in which the 1H delay unit 1
7, coefficient weighting section 21, addition section 20, MC coefficient setting section 22
It consists of.

A group of 1H delay units 17, as shown in FIG. 18, each of the signal series of the current field and the previous field.
Generate the signal sequence of the field. And the coefficient weighting unit 21
Is the coefficient V1 (j) shown in equation 2 for the current field signal sequence (corresponding to V1, 1, 2, 3, 4, 5 in the drawing) and the coefficient V1 (j) for the previous field signal sequence. In this case, the coefficient V2 (j) (corresponding to 1, 2, 3, 4, 5 of V2 in the drawing) shown in Expression 3 is weighted. Adder 20
Adds these coefficient-weighted signals to obtain an interpolated signal MCIP by the motion correction processing shown in Expression 1 on the output. In addition,
Although not explicitly shown in the drawings, the same configuration is required for each of the luminance signal and the two color difference signals (a single signal can be formed by time division multiplexing).

The MC coefficient setting unit 22 calculates the motion vector PV
The coefficient weighting unit 21 calculates the coefficient V1 shown in Equations 2 and 3 based on
(j) Generate a signal Cef for setting V2 (j).

FIG. 5 shows an outline and an example of the characteristics when the motion vector PV is detected in two frame periods. In the chart, PVy is the vertical component of the motion vector PV, y is the amount of vertical motion when converted to motion per field, q and r are the corresponding integers and decimals, and the relationship of y = 2h (q + r) Have Also, 2
h is the scanning line interval in the interlaced scanning system, and the scanning line interval is h in the sequential scanning system. V1 (j) and V2 (j) are coefficient values for the signal series of the current field and the signal field of the previous field, which are calculated by equations 2 and 3, respectively. Also, the V1 (0) scan line is the position of the current field scan line taking the coefficient V1 (0), V2 (0)
The (0) scanning line indicates the position of the scanning line in the field before taking the coefficient V2 (0), and this positional relationship is shown in the remarks.

First, the PVy moves downward from 0 to 16,
That is, in the downward movement of 0 to 4 h per field, q increases in increments of 0 to 2 and r in increments of 0.125, and is required for the motion correction processing by the calculations of equations 2 and 3 using q and r. Coefficient V1
(j) and V2 (j) are set. When the motion is h or 3h, the signal sequence obtained by moving the position of the scanning line of the previous field by the motion vector has the same phase as the signal sequence of the scanning line of the current field. Therefore, the precondition of GST (the signals of the two sequences have different phases) is broken, and it becomes impossible to generate the signal of the interpolation scanning line. For this reason, when the movement is h or 3h, as indicated in the chart by VT, VT
Generate an interpolation signal by filter processing.

On the other hand, the upward movement of PVy from -1 to -15,
That is, in the upward movement of 0 to -3.75 h per field, q decreases in -1 to -2 and r decreases in steps of 0.125.
The coefficients V1 (j) and V2 (j) necessary for the motion correction processing are set by the calculations of Equations 2 and 3 using r. If the movement is -h or -3
At the time of h, the signal sequence obtained by moving the position of the scanning line of the previous field by the motion vector has the same phase as the signal sequence of the scanning line of the current field. Therefore, as described above, the interpolation signal is generated by the VT filter processing specified as VT in the chart.

The values of the motion vector PVy can define parameters (q, r, scanning lines of V2 (0), coefficients V1 (j), V2 (j)) necessary for the motion compensation processing by GST. Therefore, this chart can be stored in a ROM or the like, and the signal Cef can be generated by table lookup processing using the motion vector PVy.

Next, the setting section and the selection section will be described with reference to FIG.

FIG. 5A shows an example of this configuration, in which the selection unit is a coefficient weighting unit 23 and an addition unit 24, and the setting unit is an interpolation mode setting unit 2.
5 and the multiplex coefficient setting unit 26.

The interpolation mode setting unit 25 outputs the motion vector PV
The interpolation mode is determined by the magnitude of the absolute value of the horizontal component PVx and vertical component PVy.
Set the signal MD. An example of this characteristic is shown in FIG.
First, when the motion vector PV is 0, it is determined to be a still area,
1 corresponding to stillness is output to MD. Also, | Pvx | <Vx
1, When | PVy | <Vy1, it is judged as a slow-moving area,
Output 2 corresponding to quasi-stationary. | PVx | is almost 0
When | Pvy | <Vy2, the vertical pan area (for example, the entire screen or the text area moves at a uniform speed in the vertical direction)
And outputs 3 corresponding to the vertical pan. On the other hand, | P
When vx |> Vx1 or | PVy |> Vy1, it is determined that the region is a normal motion region, and 4 corresponding to a moving image is output.

The multiplexing coefficient setting unit 26 generates a multiplexing control signal CT for designating coefficient values W1, W2, W3 for weighting the interpolation signal by the selecting unit based on the interpolation mode signal MD and the prediction error MVE. Generate. FIG. 2B shows an example of this characteristic.

When MD = 1, an interpolation signal suitable for a still image
W1 = 1.0, W2 and W3 are set to 0 so that the interpolation signal is generated by SIP.
Is output.

When MD = 2 or MD = 3, it is an area of motion in which the reduction of resolution and the occurrence of flicker interference cannot be avoided in the prior art, and in order to avoid this, an interpolation signal by interpolation processing for motion correction is used. W1 is 0, W2 so that
Generates a multiplex control signal CT where k is 1 and W3 is 1-k. Here, the coefficient k is 0 to 1 depending on the magnitude of the prediction error MVE of the motion vector.
Set the value of. That is, as shown in the remarks in the figure,
When MVE is smaller than the set value th and MVE <th, it is determined that the accuracy of the motion vector is high, and k is set to 1.0.
Use IP. On the other hand, when the prediction error MVE is in the range of th <MVE <th1, the accuracy of the motion vector is not very good, and there is a possibility that the image quality may be degraded due to this. Therefore, as shown in the following Expression 5, the coefficient k is set so that the value of the coefficient k gradually decreases to 0 as the prediction error increases.

K = 1.0− (MVE−th) / (th1−th) Equation 5 When the prediction error MVE is equal to or more than the set value th1, the accuracy of the motion vector is poor and the image quality is degraded due to this. Judge, set k = 0, and prohibit the use of the interpolation signal by this motion correction processing.

When MD = 4, it is determined that there is no problem even if the conventional interpolation signal for moving images is used, and W1 and W2 are 0 and W3.
= 1.0, and outputs a multiplex control signal CT so as to generate an interpolation signal with an interpolation signal VTIP suitable for a moving image.

Returning to FIG. 7A, the coefficient weighting unit 23 of the selection unit
Weights the interpolation signals SIP, MCIP, and VTIP by the coefficients W1, W2, and W3. Then, the addition unit 24 adds these, and generates a signal IP of the interpolation scanning line. Although not explicitly shown in the drawings, similar components are required for the luminance signal and the color difference signal.

Finally, the speed doubling unit will be described with reference to the drawing of FIG. Scan line signal S3 transmitted by interlaced scanning
And the signal IP of the interpolated scanning line is a write operation having a period of one scanning line period fh of the interlaced scanning system.
1, 27-2. On the other hand, the read operation from the memory unit is the RD operation shown in the drawing, and is performed alternately every one scanning line period fh / 2 cycle of the sequential scanning system. The multiplexing unit 28 performs a time-series multiplexing process on these signals, and obtains a signal S4 obtained by converting the signals into progressively scanned signals by interpolation processing for motion correction. Although not specifically shown in the drawings, the same configuration is required for the luminance signal and the color difference signal.

As described above, according to the first embodiment of the present invention, it is possible to realize a high-speed and high-resolution image signal double-speed conversion circuit which cannot be achieved by the conventional technique.
A remarkable effect is obtained in improving the image quality of V, PDP, LCD, and the like.

Next, a second embodiment of the present invention will be described with reference to FIG. The present embodiment is suitable for reducing the circuit scale, and is configured by replacing the VT interpolation generator 5 in the first embodiment with a moving image interpolation generator 29. In addition,
Other blocks are the same as those of the first embodiment,
The description of the configuration and operation is omitted.

In the present embodiment, the moving image interpolation generation unit 29
An interpolated signal MIP suitable for a moving image is generated by signal processing within the field. That is, the interpolation signal MIP is generated with the average value of the signals of the scanning lines above and below the same field as the interpolation scanning line. For this reason, the configuration is simpler than that of the VT interpolation generation unit 5 of the first embodiment, and the circuit scale can be reduced.

Next, a third embodiment of the present invention will be described with reference to FIGS. This embodiment is also suitable for reducing the circuit scale.

FIG. 9 shows an example of this block configuration, which is realized by a configuration in which the stationary interpolation generation unit 4 in the first embodiment is omitted. Then, the selection unit 30 and the setting unit 31 perform two interpolation signals VTIP and MCI in accordance with the motion vector PV and the prediction error MVE.
By changing the mixing ratio of P, a signal IP of the interpolation scanning line is generated.

FIG. 10 shows an example of one characteristic in the setting section 31. As in the first embodiment, the interpolation mode signal MD is set according to the magnitudes of the absolute values of the horizontal component PVx and the vertical component PVy of the motion vector PV. First, when the motion vector PV is 0, 1 corresponds to stationary, when | Pvx | <Vx1 and | PVy | <Vy1, 2 is quasi-static, and when | PVx | is almost 0 and | Pvy | <Vy2, Vertical pan 3, | P
When vx |> Vx1 or | PVy |> Vy1, 4 corresponding to the moving image is set.

The interpolation mode signal MD and the prediction error MVE
, A multiplex control signal CT that specifies coefficient values W1 and W2 to be weighted is generated.

When MD = 1, as described above with reference to FIG.
The interpolation processing of the motion correction is performed when PVy = 0 in FIG.
This corresponds to a case where both r are 0. Then, the coefficient V2 (0) is 1, the other coefficients V2 (j) and V1 (j) are all 0, and the interpolation signal MCI
P is the signal of the scanning line A in the previous field, that is, the same signal as the interpolation signal of the static interpolation generation unit 4 in the first embodiment. Therefore, as this produces an interpolation signal,
The multiplex control signal CT that satisfies W1 = 1.0 and W2 = 0 is output.

When MD = 2 or MD = 3, it is an area of motion in which the reduction of resolution and the occurrence of flicker interference cannot be avoided in the prior art. W1 is k, W2 so that
Generates a multiplex control signal CT with 1-k. Where the coefficient k
Sets a value of 0 to 1 according to the magnitude of the motion vector prediction error MVE, as in the first embodiment. That is, M
When VE is less than the set value th and MVE <th, k = 1.0, prediction error MVE
However, in the range of th <MVE <th1, the coefficient k is set with a characteristic that the value of the coefficient k gradually decreases to 0 as the prediction error increases, as shown in Expression 5. In addition, the prediction error MVE is equal to the set value th.
If it is 1 or more, set k = 0.

When MD = 4, it is determined that there is no problem even if the conventional interpolation signal for moving images is used, and W1 = 0 and W2 = 1.0.
Set to.

The selector 30 weights the coefficient W1 to the interpolation signal MCIP and the coefficient W2 to the interpolation signal VTIP and adds them to generate the signal IP of the interpolation scanning line.

The configuration and operation of each block other than the above are the same as those of the first embodiment, and a description thereof will be omitted.

Next, a fourth embodiment of the present invention will be described with reference to FIGS.

FIG. 11 shows an example of this block configuration, which is realized by a configuration in which the stationary interpolation generation unit 4 in the second embodiment is omitted. Then, the selection unit 30 and the setting unit 32 perform two interpolation signals VTIP and MCI in accordance with the motion vector PV and the prediction error MVE.
By changing the mixing ratio of P, a signal IP of the interpolation scanning line is generated.

FIG. 12 shows an example of one characteristic in the setting section 32. As in the third embodiment, the interpolation mode signal MD is set according to the magnitude of the absolute value of the horizontal component PVx and the vertical component PVy of the motion vector PV. First, when the motion vector PV is 0, 1 corresponds to stationary, when | Pvx | <Vx1 and | PVy | <Vy1, 2 is quasi-static, and when | PVx | is almost 0 and | Pvy | <Vy2, Vertical pan 3, | P
When vx |> Vx1 or | PVy |> Vy1, 4 corresponding to the moving image is set.

Then, the interpolation mode signal MD and the prediction error MVE
, A multiplex control signal CT that specifies coefficient values W1 and W2 to be weighted is generated.

When MD = 1, similarly to the third embodiment, W1 =
The multiplex control signal CT that makes 1.0 and W2 = 0 is output.

When MD = 2 or MD = 3, W1 is
k and W2 generate a multiplex control signal CT with 1-k. Here, the coefficient k is the motion vector prediction error M, as in the third embodiment.
Set a value between 0 and 1 according to the size of VE. That is, MVE
Is less than the set value th and MVE <th, k = 1.0, MVE is th <MVE
Equation 5 is set in the range of <th1, and k = 0 when MVE is equal to or greater than the set value th1.

When MD = 4, it is determined that there is no problem even if the conventional interpolation signal for moving images is used, and W1 = 0 and W2 = 1.0.
Set to.

The selecting section 30 weights the coefficient W1 to the interpolation signal MCIP and the coefficient W2 to the interpolation signal MIP and adds them to generate the signal IP of the interpolation scanning line.

The configuration and operation of each block other than the above are the same as those of the previous embodiments, and the description will be omitted.

As described above, according to the second to fourth embodiments of the present invention, it is possible to realize a high-speed and high-resolution image signal double-speed conversion circuit with a reduced circuit scale. This has a remarkable effect on cost reduction and high image quality of multimedia TV, PDP, LCD and the like.

In the above-described embodiment, the detection of the motion vector PV necessary for the motion correction processing is performed only by the signal S1 and the signal S2 obtained by delaying the signal S1 by a plurality of frame periods. However, in recent years, as a means for efficiently transmitting an image signal, a scheme for efficiently encoding an image signal by predictive coding of motion compensation has been developed, and is expected to be widely used in fields such as digital broadcasting in the future. . In the predictive coding of the motion compensation, a position of a motion vector is moved to generate a signal of a predicted frame, and a difference component between the predicted frame signal and the signal of the current frame and a motion vector are coded and transmitted. On the decoding side, the original frame is obtained from the decoded difference component and the motion vector.
Decode the system signal.

In the following, an embodiment of the present invention utilizing the motion vector of this motion compensation predictive coding will be described.

FIGS. 13 and 14 show a fifth embodiment of the present invention, in which the motion vector is utilized in the first embodiment.

FIG. 13 shows an example of this configuration, which is realized by a configuration in which a motion vector converter 33 is added to the first embodiment.

The motion vector conversion unit 33 converts the coded motion vector MVD obtained by decoding the motion compensated prediction coded video signal into a plurality of frame periods same as the motion vector used in the motion compensation interpolation processing. And performs a vector conversion process for converting the motion vector into a motion vector, and outputs this as a reference vector VR.

FIG. 14 shows an outline of the operation of this vector conversion. FIG. 1A is a schematic diagram of conversion of a motion vector used in P-picture coding in motion compensation prediction coding.
In P picture encoding, a frame signal indicated by symbols I and P of an image signal sequence is encoded by unidirectional prediction. The P vector for generating the predicted frame signal is a motion vector between a plurality of frames (three frames in the drawing).
On the other hand, in the present invention, for example, a motion vector in two frame periods is used. Therefore, a conversion vector is generated by a vector conversion process of multiplying the P vector by N / M (2/3 times in the drawing).

On the other hand, in the B picture encoding, the frame of the symbol B is encoded by bidirectional prediction as shown in FIG. Then, a motion vector indicated by a solid line and a dotted line is used as the B vector. So, for example, one of these frames
A conversion vector is generated by a vector conversion process of multiplying the motion vector in the time period by N.

Returning to FIG. 13, the motion vector detecting section 2
In setting the representative vector of the target block, in addition to the reference vector RMV shown in the previous embodiment, this reference vector VR
Is used to set the representative vector.

Note that the subsequent configuration and operation are the same as those of the first embodiment, and therefore, description thereof will be omitted.

Next, a sixth embodiment of the present invention is shown in FIG. This embodiment is realized by a configuration in which the motion vector conversion unit 33 described in the fifth embodiment is added to the second embodiment.

FIG. 16 shows a seventh embodiment of the present invention. This embodiment is realized by a configuration in which the motion vector conversion unit 33 described in the fifth embodiment is added to the third embodiment.

FIG. 17 shows an eighth embodiment of the present invention. This embodiment is realized by a configuration in which the motion vector conversion unit 33 described in the fifth embodiment is added to the fourth embodiment.

The configurations and operations of the above-described sixth to eighth embodiments can be easily understood from the above description, and the description is omitted.

According to the fifth to eighth embodiments described above, the detection of the motion vector used for the motion compensation processing can be performed with higher accuracy, and the image quality, resolution and resolution can be improved.
A double-speed image signal conversion circuit with a reduced circuit scale can be realized, and the cost of the circuit can be reduced and multimedia TV, PDP,
A remarkable effect is obtained in improving the image quality of an LCD or the like.

Next, an embodiment in which the IP conversion device according to the present invention is applied to a television receiver will be described with reference to FIG.

In FIG. 19, the TV tuner section demodulates the terrestrial broadcast signal TV into a baseband television signal. The BS tuner section demodulates the satellite broadcast signal BS into a television signal in a baseband. In addition,
The terrestrial broadcast signal and the satellite broadcast signal may be analog or digital. The analog decoder performs a predetermined demodulation process on an analog television signal. Demodulate the luminance signal and the color difference signal. The digital decoder performs a predetermined decoding process on a digital television signal, and demodulates a luminance signal and a color difference signal. The external input signal Ex is a luminance signal and a color difference signal of a package medium such as a VCR or a PC image. The switch section selects these demodulated signals and external input signals.

The IP converter converts the interlaced scanning signal sequence into a progressive scanning signal sequence by the method described above. The image format conversion unit performs format conversion processing such as frame rate conversion, enlargement / reduction of image size, conversion of aspect ratio, or conversion of the number of scanning lines.
The display unit performs image display in the form of sequential scanning. A CRT, PDP, LQD, DMD, or the like is used for the display unit.

According to the embodiment of the television receiver, the first complementary signal generated by inter-field signal processing suitable for a still image, a slow motion (for example, 2 lines per field, or 2 is smaller than the original signal) A second complementary signal generated by a motion correction process suitable for an image of (motion), or a third interpolation signal generated by a vertical / temporal filter suitable for a moving image or signal processing in a field, according to the motion of the image. Since the scan conversion is selectively performed, deterioration of the vertical resolution and flicker disturbance can be prevented, and high-quality and high-resolution video can be provided.

In the present invention, the accuracy of the motion vector is evaluated based on the magnitude of the prediction error of the motion vector required for the motion correction process, and when the accuracy is low, a complementary signal (second complementary signal) of the motion correction process is used. To perform signal processing to limit the use of, scan conversion can be performed with the best complement signal for the input image,
It is possible to provide a video with higher image quality and higher resolution than the conventional technology.

[0094]

According to the present invention, it is possible to reduce the resolution and to prevent the occurrence of flicker disturbance in the prior art,
A multi-speed image signal double-speed conversion method and circuit capable of performing interlaced to sequential scan conversion with high resolution and no flicker interference characteristics can be realized. Multimedia TV, PDP, LCD
This has a remarkable effect on high image quality.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an example of a first embodiment of the present invention.

FIG. 2 is a diagram illustrating a configuration example of a motion vector detection unit and an accuracy verification unit according to the first embodiment;

FIG. 3 is a diagram illustrating a configuration example of a stationary interpolation generation unit and a VT interpolation generation unit according to the first embodiment.

FIG. 4 is a diagram illustrating a configuration example of an MC interpolation generation unit according to the first embodiment;

FIG. 5 is a diagram illustrating an example of coefficient setting characteristics of an MC coefficient setting unit.

FIG. 6 is a diagram illustrating a configuration example of a setting unit and a selection unit according to the first embodiment;

FIG. 7 is a diagram illustrating a configuration example of a speed-doubling unit according to the first embodiment;

FIG. 8 is a block diagram showing an example of the configuration of a second embodiment of the present invention.

FIG. 9 is a diagram illustrating an example of a block configuration according to a third embodiment of the present invention.

FIG. 10 is a diagram illustrating a characteristic example of a setting unit according to the third embodiment.

FIG. 11 is a diagram illustrating an example of a block configuration according to a fourth embodiment of the present invention.

FIG. 12 is a diagram illustrating a characteristic example of a setting unit according to a fourth embodiment.

FIG. 13 is a block diagram showing an example of a fifth embodiment of the present invention.

FIG. 14 is an operation schematic diagram of a motion vector conversion unit in the fifth embodiment.

FIG. 15 is a diagram illustrating an example of a block configuration according to a sixth embodiment of the present invention.

FIG. 16 is a block diagram showing an example of a seventh embodiment of the present invention.

FIG. 17 is a diagram illustrating an example of a block configuration according to an eighth embodiment of the present invention.

FIG. 18 is a schematic diagram of the principle of motion correction scanning line interpolation.

FIG. 19 is a view showing a configuration example of a TV device according to the present invention.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Frame delay part, 2 ... Motion vector detection part, 3 ... Accuracy verification part, 4 ... Stationary interpolation generation part, 5 ... VT interpolation generation part,
6: MC interpolation generator, 7, 31, 32 ... Setting unit, 8, 3
0: selection unit, 9: delay unit, 10: double speed unit, 11: control unit, 12: block motion vector detection unit, 13: singular vector correction unit, 14, 27: memory unit, 15: mini-block division search unit , 16: prediction error calculation section, 17: 1H delay section, 18: 260H delay section, 19, 21, 23 ... coefficient weighting section, 20, 24 ... addition section, 22 ... MC coefficient setting section,
25: interpolation mode setting unit, 26: multiplex coefficient setting unit, 28
... Multiplexing unit, 29. Moving image interpolation generating unit, 33.

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Mitsuo Nakajima 292 Yoshida-cho, Totsuka-ku, Yokohama-shi, Kanagawa Prefecture Inside the Multimedia Systems Development Division of Hitachi, Ltd. (72) Inventor Yasutaka Tsuru Yoshida-cho, Totsuka-ku, Yokohama-shi, Kanagawa 292 Hitachi Media, Ltd.Multimedia System Development Division (72) Inventor Takaaki Matino 292 Yoshida-cho, Totsuka-ku, Yokohama-shi, Kanagawa Prefecture Hitachi, Ltd.Video Information Media Division (72) Inventor Haruki Takada Kanagawa 292 Yoshida-cho, Totsuka-ku, Yokohama-shi, Japan Within the Hitachi, Ltd. Video Information Media Division (72) Inventor Kazuo Ishikura 5-2-1, Josuihoncho, Kodaira-shi, Tokyo System LSI Development Center, Hitachi, Ltd. F-term (reference) 5C063 BA03 BA08 BA09 BA10 BA12 CA01 CA05 CA07 CA38

Claims (6)

[Claims] An image signal double-speed scanning conversion circuit for generating a signal of a scanning line dropped by interlaced scanning by interpolation signal processing and converting the signal into a sequential scanning signal, wherein a first interpolation signal is generated by signal processing between fields. , A MC interpolation generator that generates a second interpolation signal by signal processing of motion compensation, and a VT interpolation generator that generates a third interpolation signal by signal processing of a vertical / time filter. A motion vector detecting section for detecting a motion vector of the image signal during a plurality of frame periods, an accuracy verifying section for calculating a prediction error of the detected motion vector, and the first, second, and third sections.
A setting unit for setting a coefficient value of the coefficient weight of the interpolation signal of the first interpolation signal for an image whose detected motion vector is substantially zero, the third interpolation signal for an image having a large motion,
In a slowly moving image, when the prediction error of the motion vector is less than a predetermined value, the ratio of the second interpolation signal is proportional to the magnitude of the prediction error when the prediction error is less than a predetermined value. Is gradually reduced and the ratio of the third interpolation signal is gradually increased. The coefficient value of the coefficient weighting is set, and the coefficient value is weighted added to the first, second, and third interpolation signals to perform interpolation scanning. A motion compensation double speed conversion circuit for an image signal, which generates a line signal. 2. An image signal double-speed scan conversion circuit for generating a signal of a scanning line skipped by interlaced scanning by interpolation signal processing and converting the signal into a progressive scanning signal, the signal processing between fields being suitable for a still image. A still interpolation generating section for generating a first interpolation signal, an MC interpolation generating section for generating a second interpolation signal suitable for a slowly moving image in signal processing for motion correction, and a signal processing in a field. A video interpolation generator for generating a third interpolation signal suitable for a moving image; a motion vector detector for detecting a motion vector of the image signal during a plurality of frame periods; and calculating a prediction error of the detected motion vector. An accuracy verification unit for performing the above-described first, second, and third
A setting unit for setting a coefficient value of the coefficient weight of the interpolation signal of the first interpolation signal for an image whose detected motion vector is substantially zero, the third interpolation signal for an image having a large motion,
In the case of a slowly moving image, when the prediction error of the motion vector has a high accuracy of less than a predetermined value, the second interpolation signal,
When the accuracy is lower than a predetermined value, the ratio of the second interpolation signal gradually decreases in proportion to the magnitude of the prediction error, and the ratio of the third interpolation signal gradually increases. A motion compensation double speed conversion circuit for an image signal, wherein the signal is adaptively set, and the coefficient value is weighted added to the first, second, and third interpolation signals to generate a signal of an interpolation scanning line. 3. An image signal double speed scan conversion circuit for generating a signal of a scanning line skipped by interlaced scanning by interpolation signal processing and converting the signal into a progressive scanning signal. An MC interpolation generating section for generating a first interpolation signal suitable for a video signal, a VT interpolation generating section for generating a second interpolation signal suitable for a moving image by signal processing of a vertical / time filter, and a plurality of frames of an image signal. A motion vector detecting unit that detects a motion vector in a time period, an accuracy verifying unit that calculates a prediction error of the detected motion vector, and the first, second, and third
A setting unit for setting a coefficient value of the coefficient weight of the interpolation signal of the first interpolation signal for an image whose detected motion vector is substantially zero, the second interpolation signal for an image having a large motion,
In the case of a slowly moving image, the first interpolation signal is used when the prediction error of the motion vector is accurate to a predetermined value or less.
When the accuracy is lower than a predetermined value, the ratio of the first interpolation signal gradually decreases and the ratio of the second interpolation signal gradually increases in proportion to the magnitude of the prediction error. An image signal motion correction double speed conversion circuit, which is adaptively set and generates a signal of an interpolation scanning line by weight-adding the coefficient value to the first and second interpolation signals. 4. An image signal double-speed scanning conversion circuit for generating a signal of a scanning line dropped by interlaced scanning by interpolation signal processing and converting the signal into a progressive scanning signal, wherein the image processing apparatus performs a motion correction signal processing to obtain a still and slowly moving image. MC generation unit for generating a first interpolation signal suitable for a video signal, a moving image interpolation generation unit for generating a second interpolation signal suitable for a moving image by signal processing in a field, and a plurality of frames of an image signal A motion vector detecting unit that detects a motion vector in a time period, an accuracy verifying unit that calculates a prediction error of the detected motion vector, and the first, second, and third
A setting unit for setting a coefficient value of the coefficient weight of the interpolation signal of the first interpolation signal for an image whose detected motion vector is substantially zero, the second interpolation signal for an image having a large motion,
In the case of a slowly moving image, the first interpolation signal is used when the prediction error of the motion vector is accurate to a predetermined value or less.
When the accuracy is lower than a predetermined value, the ratio of the first interpolation signal gradually decreases and the ratio of the second interpolation signal gradually increases in proportion to the magnitude of the prediction error. An image signal motion correction double speed conversion circuit, which is adaptively set and generates a signal of an interpolation scanning line by weight-adding the coefficient value to the first and second interpolation signals. 5. A motion vector detecting apparatus comprising: a motion vector converting unit for generating a reference vector in a vector conversion process of motion vector information used in predictive coding of motion compensation; 10. The circuit of claim 6, wherein the detection is performed. 6. An input section for inputting an image signal of interlaced scanning, an operation conversion section for converting the image signal from an interlaced scanning to an image signal of sequential scanning, and a display for displaying an output of the search conversion section for the image signal. A scanning interpolation unit for generating a first interpolation signal by signal processing between fields; and a MC interpolation unit for generating a second interpolation signal by signal processing for motion compensation. A generator, a VT interpolation generator that generates a third interpolation signal by signal processing of a vertical / temporal filter, a motion vector detector that detects a motion vector of the image signal in a plurality of frame periods, An accuracy verification unit for calculating a prediction error of the motion vector, and the first, second, and third
A setting unit for setting a coefficient value of the coefficient weight of the interpolation signal of the first interpolation signal for an image whose detected motion vector is substantially zero, the third interpolation signal for an image having a large motion,
In a slowly moving image, when the prediction error of the motion vector is less than a predetermined value, the ratio of the second interpolation signal is proportional to the magnitude of the prediction error when the prediction error is less than a predetermined value. Is gradually reduced and the ratio of the third interpolation signal is gradually increased. The coefficient value of the coefficient weighting is set, and the coefficient value is weighted added to the first, second, and third interpolation signals to perform interpolation scanning. A television device for generating a line signal.