JPH07154754A - Movement correction video signal processing system - Google Patents

Abstract

PURPOSE: To prevent the selection of an inaccurate moving vector at the time of selecting a vector by testing a group of moving vectors including a zero moving vector and selecting a moving vector having an weighted correlation value indicating the highest correlation. CONSTITUTION: A zero moving vector and three vectors v1 to v3 are respectively supplied to processing units 600, 610, 620, 630. In response to an address from an address deviation calculator 640 in the unit 600, each of RAMs 650, 660 supplies an array of pixel values expressing test blocks t.b to a block checker and comparator 670. The comparator 670 mutually compares pixel values existing on the corresponding positions of two test blocks, i.e. the luminance components of the pixel values, and calculates an absolute luminance difference between respective pixels. In order to indicate a synthetic correlation between two test blocks, the sum SAD of all absolute luminance differences is generated and a low SAD value of measured moving vectors indicates high correlation between test blocks specified by moving vectors.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a motion compensation video signal processing system. Motion compensated video signal processing is used in devices such as television standard conversion, film standard conversion and video-to-film standard conversion. British Published Patent Application GB-A-22317
In motion compensated television standard converters such as the converter described in No. 49, a pair of consecutive input images is processed to generate a plurality of sets of motion vectors representing the motion of the images between the input image pairs. . Since the process is performed on individual blocks of video, each motion vector represents the inter-video motion of the contents of the respective block.

Each set of motion vectors is fed to a motion vector reducer, where a subset of the set of motion vectors for each block is derived. This subset containing zero motion vectors is then passed to a motion vector selector, where one of the subsets of motion vectors is assigned to each pixel in each video block. The motion vector selected for each pixel is provided to a motion compensated interpolator, which operates to interpolate the continuous output image taking into account the motion between the input images.

Since vector selection is used to interpolate one output pixel from two input images, one motion vector is selected from a subset of motion vectors (eg, four motion vectors) supplied by a vector subtractor. Used for. The vector selection process involves detecting the degree of correlation between the test blocks of the two input images pointed to by the tested (tested) motion vector. The motion vector with the highest degree of correlation between test blocks is selected for use in interpolating the output pixels.

Problems can arise when vector selection is performed on an input image with large noise content. In this case, an incorrect motion vector may be selected due to a pseudo-apparent correlation between noise regions (portions) in the input video. A motion-compensated video signal processing device operating under such a condition may have a drastic malfunction when a certain level of image noise arrives, and its performance may be worse than that of a device without motion compensation. is there.

[0005]

SUMMARY OF THE INVENTION An object of the present invention is to prevent an incorrect motion vector from being selected due to a false correlation between noise regions in an input image when selecting a vector in motion compensation video signal processing. However, it is to prevent the performance of the motion compensation video signal processing device from deteriorating when the video noise is large.

[0006]

SUMMARY OF THE INVENTION A motion compensated video signal processing apparatus according to the present invention comprises means for generating a plurality of motion vectors representing motion of a video between input video pairs of an input video signal, and zero video. Means for testing a group of motion vectors from the group of motion vectors, including zero motion vectors indicating motion, for use in interpolating output pixels of the output video, the means for testing comprising: A means for detecting a correlation value indicating the correlation degree between the test blocks of the input video pointed to by the test motion vector, and the correlation degree indicated by the correlation value for the zero motion vector is the correlation value for other motion vectors in the group. Means for weighting the correlation value so as to increase as compared with the correlation degree represented by
Selecting a motion vector having a weighted (weighted) correlation value showing the highest correlation from the group of motion vectors.

In the motion compensated video signal processing device according to the invention, the result of the vector selective correlation test obtained for the zero motion vector is weighted compared to the other motion vectors in the group. This should be chosen to have a zero motion vector at the time of motion vector selection, unless the other vectors in the subset of motion vectors show better inter-test block correlation, under other conditions being equal. It means that.

Since the vector selection process leans toward zero motion vector selection, the motion compensated video processor returns to a performance that is not worse than a non-motion compensated device when noisy input images prevent reliable motion vector selection. be able to.

The detecting means may include means for detecting an absolute luminance difference between pixel pairs at corresponding positions in a test block, and means for summing the absolute luminance differences to generate a correlation value. In this case, the weighting means may include means for multiplying the correlation value for the zero motion vector by a predetermined constant value smaller than one. That is, the weighting acts to reduce the correlation value for the zero motion vector compared to the correlation values for the other motion vectors (to show a greater degree of correlation).

The weighting means may include means for multiplying a correlation value for one or more motion vectors other than the zero motion vector in the group by a weight value larger than one, respectively.

Alternatively, additive (or subtractive) weighting may be used. For this purpose, the weighting means may comprise means for subtracting a predetermined constant value from the correlation value for the zero motion vector. Similarly, the weighting means may include means for adding weighting values to the correlation values for one or more motion vectors other than the zero motion vector in the group.

In another preferred embodiment, the weighting means reduces the correlation value so that the correlation degree indicated by the correlation value for each motion vector in the group is reduced by an amount proportional to the magnitude of the motion vector. It operates to add weight. This means that large motion vectors should produce better correlation between test blocks than small motion vectors, if all other factors are equal in order to be selected.

The present invention, in a second aspect, provides a motion compensated video signal processing method. The method comprises the steps of generating a plurality of motion vectors representing motion of a video between input video pairs of an input video signal, and outputting an output video from a group of motion vectors including a zero motion vector indicating motion of a zero video. Testing the group to select one motion vector for interpolation of pixels, the testing step comprising: between test blocks of an input video pointed to by the motion vector under test (under test). Detecting a correlation value indicative of the correlation value of the correlation vector such that the correlation value indicated by the correlation value for the zero motion vector is greater than the correlation values indicated by the correlation values for the other motion vectors in the group. Weighting the values, the motion with the weighted correlation value showing the highest correlation from the group of motion vectors Including each process of selecting a vector.

The fine details moving the vector selection process in front of the static background (one of the test blocks used for the correlation test)
Incorrect vector may be selected for interpolation of moving parts. This is because the correlation between the regions around the moving part in the two input images is completely different, and this difference influences the result of the correlation test. An example of this potential problem is shown in FIGS.

1a and 1b and 2a and 2b show problems in the vector selection process for the image of the automobile. In the vector estimation, at least a zero motion vector and a horizontal motion vector are respectively specified as the motion vector background and the motion of the automobile. For the body part, the correlation test performed when selecting the vector should be able to easily select the "car" motion vector. However, for items such as car radio antennas, the test blocks 10, 20 (a, b in FIG. 1) pointed to by the “car” motion vector dominate over various parts of the video background. Depends on. The correlation of these test blocks is low. In contrast, the correlation between the test blocks 30, 40 (a, b in FIG. 2) pointed to by the zero motion vector is high. This is due to the fact that these test blocks represent the same part of the video background and slightly differ in that one of the test blocks contains a part of the car radio antenna. Therefore, the result of the correlation test for the zero motion vector shows a greater correlation than the result for the "car" motion vector, so the zero motion vector is incorrectly selected for use in interpolating some of the car radio antennas. .

To avoid this problem, the size of the test block may be reduced, but this method will make the vector selection process susceptible to noise in the input video.

The present invention provides a motion-correction video signal processing device having the following configuration from the third aspect. The apparatus includes means for generating a plurality of motion vectors representative of motion of a video between input video pairs of an input video signal and selecting one motion vector from the group of motion vectors for use in interpolating output pixels of an output video. Further comprising means for testing the group and a motion compensation interpolator for interpolating output pixels from the input video pair of the input video signal using the selected motion vector, the means for testing comprising: Video enhancement means for enhancing at least the video content of the test blocks of the input video pointed to by, means for detecting the degree of correlation between the video enhanced test blocks, and the highest correlation between the respective video enhanced test blocks. Means for selecting a motion vector having a degree.

According to this aspect of the invention, video enhancement is used to enhance at least the test blocks of the input video pair. In this way, it is possible to increase the visibility of the minute portion such as the above-mentioned car radio antenna or the importance in the test block. Since the fine portion is emphasized, the influence on the result of the correlation test performed at the time of vector selection becomes large. This increases the possibility that an accurate motion vector is selected in the vector selection process, for example, for a small minute object that moves with respect to a still background.

Video enhancement may be performed before or after deriving the test block from the input video.

In a preferred embodiment, the image enhancing means comprises a two dimensional high pass spatial filter. However, although a high-pass filter can be used to enhance the fine details in the test block, the filter also makes the image noise more noticeable, thus degrading the overall system performance. Thus, other implementations use vertical Sobel and horizontal Sobel video filters (these filters perform highpass filtering in one direction and lowpass filtering in the other direction). And means for detecting an output by the vertical Sobel video filter or the horizontal Sobel video filter, which is greater than a predetermined degree of detail around one or more pixels in the test block, and in response to the detection means,
Means for replacing one or more pixels of the test block with pixels of a predetermined intensity. Pixels with a given brightness are
It should be a white pixel.

In another embodiment, the image enhancement means comprises a vertical Sobel image filter, a horizontal Sobel image filter, and a vertical Sobel image filter, which indicates less than a predetermined degree of detail around one or more pixels in the test block. A second detecting means for detecting an output by the horizontal Sobel image filter, and a test block 1 in response to the second detecting means.
Means for replacing the above pixels with pixels having a second predetermined brightness.

In this case, the pixel having the second predetermined brightness is preferably a black pixel.

In yet another specific configuration, the image enhancing means is
Means for detecting an output by a vertical Sobel video filter or a horizontal Sobel video filter indicating a degree of detail around one or more pixels in the test block, and one or more pixels of the test block responsive to the detection means And a means for adding a predetermined brightness component to the brightness component.

This aspect of the invention may be combined with weighting the correlation values of zero motion vectors. For this purpose, the group of motion vectors includes a zero motion vector indicating the motion of a zero video, and the apparatus determines the correlation between the video enhancement test blocks of the input video pointed to by the motion vector under test. Means for detecting the correlation value represented,
Means for weighting the correlation value such that the degree of correlation indicated by the correlation value for the zero motion vector is increased relative to the degree of correlation indicated by the correlation values for the other motion vectors in the group, and The selecting means preferably operates to select a motion vector having a weighted (weighted) correlation value indicating the highest degree of correlation.

The detecting means detects the absolute luminance difference between the pixel pairs at corresponding positions in the video enhancement test block, and means for summing the absolute luminance differences to generate a total absolute difference (SAD) value. It should include and.

According to a fourth aspect, the present invention provides a motion compensated video signal processing method. The method comprises the steps of generating a plurality of motion vectors representing motion of a video between an input video pair of an input video signal, and selecting a motion vector used to interpolate output pixels of an output video from a group of motion vectors, Testing the group and interpolating an output pixel from an input image pair of an input video signal using a selected motion vector, the testing step comprising: inputting an input pointed to by a motion vector under test. Each process of video enhancement of at least the video content of the test block of the video, detection of the correlation between the video enhancement test blocks, and selection of the motion vector having the highest correlation between the respective video enhancement test blocks. Is included.

FIG. 3 shows another problem that occurs when a fine object moves in front of a very simple background (in the figure, the O-shaped background moves from field to field from left to right). In this case, the correlation test performed with vector selection may give a very good match between the two regions of the simple background. In fact, the match obtained between two simple regions often looks better than the match obtained between corresponding two parts of a fine object. Since the motion vector with the highest degree of correlation between test blocks is selected for use in interpolation, a very good match between simple regions is
This causes the result that an incorrect vector is selected in preference to a correct vector representing the movement of a fine object. The result of this error appears as a visible gap (actually a portion of the background) in the fine object.

The present invention provides a motion compensation video signal processing device having the following configuration in the fifth aspect. The apparatus includes means for generating a plurality of motion vectors representing motion of a video between input video pairs of an input video signal, and a group of motion vectors for selecting a motion vector used to interpolate output pixels of an output video. A means for testing, the means for testing generating means for detecting a degree of correlation between test blocks of an input image pointed to by a motion vector under test and an analysis value indicating a degree of image detail of the test block. And means for selecting one motion vector from the group of motion vectors according to the degree of correlation between the test block pointed to by the motion vector and the analysis value for the test block pointed to by the motion vector. .

In this aspect of the invention, parsed values are generated to represent the level of detail in the test blocks and are used to influence the vector selection process. In this way, the correlation found between the detailed test blocks will be
Greater weight can be given to correlations found between simple (non-detailed) test blocks. This can alleviate the problem of error vector selection described above.

There are various methods for detecting the analysis value. For example, the output of one or more Sobel filters may be utilized. In a preferred embodiment, the means for generating the analytic value comprises means for detecting the difference between the highest intensity component of the pixel in the test block and the lowest intensity component of the pixel in the test block.

In another implementation, the means for generating the analytic value includes means for summing the brightness differences between adjacent pixel pairs within each test block. The summing means adds, for each pixel in the test block, the higher of the luminance differences between the pixel and the vertically adjacent pixel, and the luminance difference between the pixel and the horizontally adjacent pixel. Including means.

The result (analysis value) of the detailed test can influence the vector selection process in various ways. In a preferred embodiment, the apparatus includes means for preventing selection of a motion vector having an analysis value indicating that the test block pointed to by the motion vector is lower than a predetermined video detail.

Alternatively, a weight may be added to the test result indicating the correlation between the test blocks. In this case, the device
A means for detecting a correlation value indicating the correlation degree between the test blocks, and the correlation value indicated by the correlation value for the motion vector according to the analysis value so that the correlation value increases as the detected video detail increases. And means for weighting
It is preferable that the selecting means performs an operation of selecting a motion vector having a weighted correlation value indicating the highest degree of correlation.

In a simple and advantageous embodiment, the means for detecting the degree of correlation are the means for detecting the absolute brightness difference between the pixel pairs at corresponding positions in the test block and the absolute brightness difference for summing the absolute difference. Means for generating a (SAD) value. In this case, the weighting means may include means for multiplying the absolute difference total value by a weighting coefficient that decreases as the detected image detail level increases.

Video enhancement can also be used in this aspect of the invention. In this case, a video enhancing means for enhancing the video content of at least the test block of the input video pointed by the motion vector to be tested is used.

It is also preferable to use the zero vector weighted feature. For this purpose, the group of motion vectors includes a zero motion vector indicating the motion of the image of zero, and the apparatus uses a correlation value indicating the degree of correlation between the test blocks of the input image pointed to by the tested motion vector. And means for weighting the correlation value such that the degree of correlation indicated by the correlation value for the zero motion vector is greater than the degree of correlation indicated by the correlation value for the other motion vector in the group. It is preferable that the selecting means performs the operation of selecting the motion vector having the weighted correlation value indicating the highest degree of correlation.

In the preferred embodiment, the input video pair comprises two consecutive fields of the interlaced input video signal. The output video may include one field of the interlaced output video signal.

Viewed from a sixth aspect, the present invention provides a motion compensated video signal processing method comprising the steps of: The method comprises the steps of generating a plurality of motion vectors representing the motion of the video between the input video pairs of the input video signal, and a group of motion vectors to select the motion vectors used to interpolate the output pixels of the output video. A step of testing, the step of testing detecting a degree of correlation between test blocks of an input image pointed to by a motion vector under test, generating an analysis value indicating a spatial frequency content of the test blocks, The steps of selecting a motion vector from the group of motion vectors according to the degree of correlation between the test block pointed to by the motion vector and the analysis value for the test block pointed to by the motion vector. .

The device according to the invention is particularly suitable for use in television standard converters.

[0040]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be specifically described below with reference to the drawings. FIG. 4 is a block diagram showing a motion compensation television standard conversion apparatus in which the present invention can be used. The apparatus is capable of input interlaced digital video signal 250 (eg, 1125).
/ 60, 2: 1 high definition video signal (HDVS) is received and output interlaced digital video signal 260 (eg,
1250/50, 2: 1 signal).

The input video signal is first provided to the input buffer packer 310. For a normal definition input signal, the input buffer packer 310 converts the video data to a high definition (16: 9 aspect ratio) format and fills black pixels as needed. For HDVS input, the input buffer packer 310 merely temporarily stores the data.

The data is input buffer packer 310.
From the matrix circuit 320, which (if necessary) uses the standard CCIR colorimetric method for the input video signal.
Recommendation 601 "(Y, Cr, Cb) colorimetric method.

The input video signal is a matrix circuit 320.
To time base conversion (TBC) and delay (delay)
It is sent to the circuit 330 and to the subsampled TBC and delay circuit 380 via the subsampler 370. The TBC and delay circuit 330 determines the time position of each field of the output video signal and uses it for interpolation of the output field, and therefore selects the input video signal of the two fields closest in time to the output field. For each field of the output video signal, the two input fields selected in the TBC circuit are properly delayed before being sent to the interpolator 340 which interpolates the output field. The control signal t indicates the time position of each output field with respect to the selected two input fields, and includes a time base conversion (TBC) and delay circuit 330 to an interpolator 340.
Is supplied to.

The sub-sampled TBC and delay circuit 380 operates similarly, except that the sub-sampler 37
The difference is that it uses spatially sub (down) sampled video supplied by 0. The TBC circuit 33
The field pair corresponding to the field pair selected by 0 is subsampled by the TBC and delay circuit 380.
Is selected from the subsampled video and used to generate the motion vector.

The TBC circuits 330 and 380 can operate according to sync signals associated with the input video signal, the output video signal, or both. If only one sync signal is supplied, the timing of the other field of the two video signals can be deterministically generated in the TBC circuits 330, 380.

The field pairs of the sub-sampled input video signal, selected by the sub-sampled TBC and delay circuit 380, are fed directly to the block matcher 3
90, correlation surface processor 400, motion vector estimator 41
0, motion vector subtractor 420, motion vector selector 4
And a motion vector post-processor 440. The pair of input fields is first sent directly to the block matcher 390 where there is a search block in the earliest of the two selected input fields and in the later of the two input fields (greater than ) Compute a correlation surface that represents the spatial correlation with the search area.

The correlation surface processor 400 generates a number of interpolated correlation surfaces from the correlation surface output from the block matching unit 390, and these are sent to the motion vector estimator 410. The motion vector estimator 410 detects the maximum correlation point in the interpolated correlation surface. (Since the original correlation surface actually represents the difference between the blocks of the two input fields, the maximum correlation point becomes the minimum point on the correlation surface. Therefore, hereinafter, it is referred to as the "minimum point".) To detect the minimum point, The points are supplemented and interpolated in the correlation surface to some extent compensate for the loss of resolution caused by using the subsampled video to generate the correlation surface. The motion vector estimator 410 generates a motion vector from the minimum point in each detected correlation surface and sends it to the motion vector subtractor 420.

The motion vector estimator 410 also performs a confirmation (confidence) test for each motion vector generated,
Check if the motion vector is well above the average data level and attach a confirmation flag to each motion vector to indicate the result of the confirmation test. The confirmation test is known as the "threshold" test and is described in GB-A-2,231, above.
749 (along with some other features of the device of FIG. 1).

The motion vector estimator 410 also performs a test to detect if each vector is fake. In this test, the correlation surface (excluding the exclusion area around the detected minimum point) is examined to detect the next minimum point. If this second minimum is not at the edge of the exclusion zone, then the motion vector derived from the first minimum is flagged as potentially false.

The motion vector subtractor 420 reduces the possible motion vector selection for each pixel of the output field before sending the motion vector to the motion vector selector 430. The output field is conceptually divided into blocks of pixels. Each of those blocks is
It has a position in the output field that corresponds to the search block in the earlier of the selected input fields. The motion vector subtractor associates a group of four motion vectors with each block of the output field, and each pixel in the block is interpolated with the selected one of the four motion vectors of the group.

A vector flagged as "fake" is requalified during vector reduction if it is the same as the unflagged vector in the nearest block.

The motion vector subtractor 420, as part of its function, generates a "correct" motion vector (ie, a motion vector that passes verification and counterfeit tests, or requalified as not counterfeit). The frequencies are counted without considering the position of the blocks of the input field used to obtain their motion vectors. The proper motion vectors are then ranked in order of decreasing frequency. The most common of the appropriate motion vectors that differ significantly from each other are classified as "wide area" motion vectors. The three motion vectors that pass the verification test are then selected for each block of output pixels and, along with the zero motion vector, are sent to the motion vector selector 430 for further processing. These three selected motion vectors are selected from the following in a predetermined priority order. (I) motion vectors generated from corresponding search blocks (“local” motion vectors), (ii) those generated from surrounding search blocks (“adjacent” motion vectors),
(iii) Wide area motion vector.

The motion vector selector 430 also receives as input the two input fields selected by the subsampled TBC and delay circuit 380 and used to calculate the motion vector. These fields are appropriately delayed and provided to the motion vector selector 430 at the same time as the vectors derived from these fields.
Motion vector selector 430 provides an output containing one motion vector per pixel in the output field. This motion vector is selected from the four motion vectors for the block provided by the motion vector subtractor 420.

The vector selection process involves the detection of the degree of correlation between the test blocks of the two input fields pointed to by the motion vector under test. The motion vector with the highest degree of correlation between test blocks is selected for use in interpolating the output pixels. The vector selector also generates a "motion flag". This flag
"Still" if the correlation between the blocks pointed to by the zero motion vector is greater than a preset threshold
Set to (no movement).

The vector post-processor 440 modifies the format of the motion vector selected by the motion vector selector 430 to represent the vertical or horizontal scaling of the image, if any, and interpolates the vector with this format changed. To the container 340. The interpolator 340 uses the motion vector to interpolate the output field from the corresponding two (non-subsampled) interlaced input fields selected by the TBC and delay circuit 330. In this case, any motion of the image represented by the motion vector currently supplied to the interpolator 340 is considered.

If the motion flag indicates that the current output pixel is within the motion portion of the image, then the pixels from the two selected fields supplied to the interpolator are the two input fields of the output field ( Depending on the time position (as indicated by the control signal t), they are combined in relative proportions. That is, closer input fields are combined in a larger proportion. When the motion flag is set to "still", the temporal weighting is fixed at 50% of each input field. The output of the interpolator 340 is sent to the output buffer 350 and is output as a high definition output signal, and is also sent to the down converter 360 and is output as a normal definition output signal 365.

The down-converter 360 enables the display of the output of the present apparatus (for example, a high definition video signal) to be monitored, transmitted or recorded using a conventional definition apparatus. . This is useful as conventional high definition devices are much cheaper and much more prevalent than high definition devices. For example, simultaneous transmission of normal and high definition video would be required for transmission over terrestrial and satellite channels, respectively.

The sub-sampler 370 has a matrix 32.
The input video fields received from 0 are spatially (vertically) sub-sampled in the horizontal and vertical directions and then supplied to a time base conversion (TBC) and delay circuit 380. The horizontal sub-sampling consists of a half-bandwidth low-pass filter for the input field (in this example of 2: 1 horizontal decimation).
Is pre-filtered and discards every other video sample along each video line, which reduces the number of samples along each video line by half.

Vertical subsampling of the input field is complicated because the input video signal 250 is interlaced. This means that the continuous line of video samples in each interlaced field is effectively split into two video lines, and the line in each field is vertically offset by one complete video line from the lines in the preceding and following fields. means.

One method of vertical sub-sampling is to perform progressive (continuous or sequential) scan conversion (producing consecutive progressively scanned video frames, each having 1125 lines), and then progressively scanning the progressively scanned frames. Vertical sub-sampling would be done, sub-sampling at a rate of two. However, efficient progressive scan conversion requires some motion compensation processing, which can adversely affect the operation of motion processor 385. In addition, real-time progressive scan conversion of high definition video signals would require exceptionally powerful and complex processing equipment.

A simpler method of vertical spatial sub-sampling is to first low-pass filter the input field vertically (to avoid potential aliasing) and then each pixel as shown in FIG. The filtering is performed so as to be effectively shifted vertically by ½ of the video line (for even fields) or upward (for odd fields). The resulting offset field is broadly equivalent to a vertically subsampled progressive scan frame at a rate of two.

Thus, in summary, as a result of the subsampling operation described above, the motion processor 385 will operate on spatially subsampled input field pairs at a rate of 2 in the horizontal and vertical directions. This reduces the processing required for motion vector estimation to 1/4.

FIG. 6 is an explanatory diagram showing a comparison of test blocks when selecting a vector. This figure shows output field 5
Two motion vectors 500 and 510 associated with the block containing the output pixel 520 at 30 are shown. For clarity of illustration, the other two motion vectors associated with output pixel 520 by motion vector subtractor 420 are not shown.

To test one of the four motion vectors associated with output pixel 520 (eg, motion vector 510), extrapolate that motion vector and output field 530 is interpolated from input field 56.
Pixel test block 540 at 0,570,
Point each one to 550. The degree of correlation between the test blocks of the pixel is determined by calculating the absolute luminance difference between pairs of pixels at corresponding positions in the two test blocks. These absolute luminance difference values are then added to form the absolute luminance difference sum (SAD) associated with the motion vector under test (tested). A high SAD value indicates low correlation between blocks in the input frame around the pixel pointed to by the motion vector, and a low SAD value indicates high correlation between those blocks. This test is provided by the motion vector subtractor 420 to the motion vector selector 430 4
This is done for each of the two motion vectors. From this test, the motion vector with the lowest SAD value is selected by motion vector selector 430 for use in interpolating output pixel 520.

FIG. 7 is a block diagram showing an example of the vector selector 430 of FIG. The four motion vectors for each block of output pixels are calculated by the motion vector subtractor 42.
The value 0 is supplied to the motion vector selector 430. These four motion vectors, namely the zero motion vector and the three other vectors v1, v2, v3 are supplied to four processing units 600, 610, 620 and 630, respectively.
Each of the processing units 600 to 630 has an address shift calculator 640 and two random access memories (RAM) 65.
With 0, 660 and block matching and comparator 670, the two RAMs store one relevant part of each of the input field pairs selected by the TBC (time base conversion) and delay circuit 380.

In each processing unit, the address shift calculator 640 receives the time shift control signal (t) generated by the TBC and delay circuit 380 along with the motion vector for that processing unit. Address shift calculator 640 generates from these two values a plurality of memory addresses for accessing the pixel test block pointed to by the motion vector and stored in RAM 650, 660. In response to the address provided by the address shift calculator, each of the RAMs 650, 660 provides an array of pixel values representing a test block (t.b.) to the block matching and comparator 670.

The block matching and comparator 670 compares the pixel value (specifically, the luminance component of the pixel value) at the corresponding position in the two test blocks. The absolute brightness difference between each pixel pair is calculated and the sum of all absolute brightness differences (SAD) is generated to show the overall correlation between the two test blocks. A low SAD value for the measured motion vector indicates a high degree of correlation between the test blocks pointed to by the motion vector.

The processing unit 630 uses the motion vector v3.
, And calculates the SAD value for the test block pointed to by the motion vector. Processing unit 630 then provides the SAD value and the identifier of vector v3 to processing unit 620.

In processing unit 620, the block matching and comparator 670 causes S to the vector v2.
An AD value is generated. The SAD value for v2 is then the SA for v3 received from processing unit 630.
It is compared with the D value. The lower of these two SAD values is the motion vector (v2 or v2) having a high degree of correlation between the test blocks pointed to by the motion vector.
3) is represented. Therefore, processing unit 620 outputs the lower SAD value 675 and the vector identifier 680 that identifies the vector from which this lowest SAD value was derived.

The processing unit 610 receives the vector v1 and calculates the SAD value from the vector. The SAD value for v1 is then compared to the lowest SAD value 675 of vectors v2, v3 provided by processing unit 620. The processing unit 610 then calculates the vector v1,
The minimum SAD value of v2 and v3 is output together with the identifier of the vector in which the SAD value is generated.

The processing unit 600 generates a SAD value for the zero motion vector, which is processed by the processing unit 61.
Compare with the current lowest SAD value of the vectors v1, v2 and v3 received from 0. From this comparison, the block matching and comparator 670 of the processing unit 600 outputs a (selected) vector identifier 690 indicating one of the four motion vectors for which the lowest SAD value was generated.

The SAD value for the zero motion vector (generated by block matching and comparator 670 of processing unit 600) is provided to comparator 700 where it is compared to a preset threshold 710. The SAD value for a zero motion vector is four motion vectors (zero, v
1, v2, v3), regardless of which one was selected for use in interpolating the output pixel.

Comparator 700 generates a motion flag 720 that is "set" when the SAD value for the zero motion vector is greater than threshold 710. When the SAD value for the zero motion vector is less than the threshold value 710, the motion flag is not set, indicating that the current output pixel is in a near static portion of the image.

FIG. 8 is a detailed block diagram of a portion (block matching and comparator 670) of the vector selector of FIG. In the figure, two test blocks (processing unit 6) pointed to by a zero motion vector
Output from the RAMs 650 and 660 in 00) is supplied to the block matching unit 672 which is a part of the block matching and comparing unit 670. Block matching device 672
Compares the pixels at corresponding positions in each of the two test blocks and produces a SAD value representing the correlation between the test blocks pointed to by the zero motion vector. This SAD value is directly output to the comparator 700,
It is also supplied to the multiplier 750.

The multiplier 750 multiplies the SAD value generated for the zero motion vector by the factor 1-δ. this is,
In order to reduce the SAD value for the zero motion vector, proportional weighting is applied. For example, if δ is equal to 0.2, the SAD value for the zero motion vector is 20%.
Decrease.

The weighted (added) SAD value for the zero motion vector output by the multiplier 750 is supplied to the comparator 674 forming a part of the block matching and comparator 670. The comparator 674, from the processing unit 610,
Current lowest SAD value and motion vectors v1, v2, v3
And an identifier indicating which of them corresponds to the current lowest SAD value. The comparator 674 compares the current lowest SAD value with the weighted SAD value output from the multiplier 750,
Zero motion vector (weighted SA for zero motion vector
D value is the current minimum SA received from processing unit 610
Lower than or equal to D value) or processing unit 61
An output 690 is generated that identifies one of the vector identifiers received from 0 (when the weighted SAD value for the zero motion vector is greater than the current lowest SAD value received from processing unit 610).

Using the apparatus of FIG. 8, weighting is applied only to SADs with zero motion vectors. Therefore,
If one of the other three (non-zero) motion vectors is to be selected, the correlation between the test blocks pointed to by one of those other three motion vectors is more than that for the zero motion vector. Must be proportionately good.

The SAD value for the zero motion vector is 1-
Instead of (or in addition to) multiplying δ (δ is less than 1), the SAD value for another motion vector (v1, v2 or v3) may be multiplied by a coefficient greater than 1. In other implementations, additive weighting can also be used. That is, the weighted value is SAD for the zero motion vector.
SA lesser than the value or for the remaining three motion vectors
Either add to the D value or use both. Also, when the SAD value of the best of the three motion vectors v1, v2, v3 (current lowest SAD) is sent from the processing unit 610 to the processing unit 600, the weight value may be added to the SAD value.

In all of the above cases where the SAD values for one or more of the three motion vectors v1, v2, v3 are weighted,
The weight may be increased according to the magnitude of each motion vector (for example, in proportion to the magnitude of the vector).
This means that a large motion vector must have better correlation between test blocks (ie, a lower SAD value) than a smaller motion vector if all other factors are the same in order to make a choice. means.

In another specific configuration of the present invention, the test blocks used for the motion vector selection are subjected to image enhancement, and the small areas having fine portions in these test blocks are enhanced. This process increases the visibility of small microscopic objects and thus has a greater impact on the block matching and block matching process performed by the comparator 670.

One of the video enhancement techniques suitable for use in the test block used for motion vector selection is to subject the test block to high frequency emphasis using a conventional two-dimensional filter device. For example, a 3 × 3 spatial filter using two-dimensional high-pass spatial filtering coefficients as shown in FIGS. 9 and 10 may be used. However, using a high-pass filter as shown in FIGS. 9 and 10, while emphasizing fine parts in the test block, makes video noise more noticeable and reduces overall system performance.

Another image enhancement technique is the so-called Sobel.
bel) The image data is corrected in response to the output of the filter. With these filters, video data is efficiently filtered by a low pass filter in one direction and a high pass filter in the other direction. According to this technique, the resulting enhanced (enhanced) image is shown in FIG.
And 10 are more noise free than similar images filtered using a two-dimensional high pass filter. Sobel filtering is also sensitive to natural image features such as horizontal and vertical lines.

FIG. 11 shows a set of coefficients for a vertical Sobel spatial filter, and FIG. 12 shows a set of coefficients for a horizontal Sobel spatial filter. The Sobel filtered version of the test block is used to control the selective permutation of pixel data, as described below.

FIG. 13 is a block diagram showing an example of a video enhancement circuit. The video enhancement circuit shown in the figure has a RAM 65 before sending a test block to the block matching and comparator 670.
It can be inserted into the device of FIG. 7 to process the test blocks output by 0,660. Alternatively, the video enhancement circuit of FIG. 13 may be used to enhance all the video to be stored in RAM 650, 660. In either case, video enhancement applies to the test block used for vector selection, not to the video where the final output pixel is generated.

In the device of FIG. 13, input video data 800 (eg, from RAM 650, 660) is provided in parallel to vertical Sobel filter 810 and horizontal Sobel filter 820. The vertical Sobel filter 810 generates, for each pixel of the input video data 800, an output dv indicating the vertical image detail at that position. Similarly, the horizontal Sobel filter 820 generates an output dh indicating the horizontal image detail level at the position.

Vertical and horizontal Sobel filters 810, 82
The zero outputs are provided in parallel to each pair of comparators. More specifically, the output dv from the vertical Sobel filter 810 is supplied to the comparators 830 and 840, and the output dh from the horizontal Sobel filter 820 is supplied to the comparators 850 and 860. Comparators 830 and 850 compare their respective inputs dv and dh with a threshold (t) 870 and produce outputs, respectively, that indicate whether dv or dh is greater than threshold 870. Similarly, the comparators 840 and 860 have their respective detailed values dv or dh.
With a negative threshold whose magnitude is equal to the threshold 870. The negative thresholds are the multipliers 88 for multiplying the threshold 870 by -1 respectively.
Generated by 0,890. Comparators 840 and 86
0 produces an output indicating whether the detailed value dv or dh is less than a negative threshold (-t), respectively.

The outputs of comparators 830 and 850 are provided to OR gate 900, whose output controls switch 910. That is, when either the horizontal detail value or the vertical detail value is larger than the threshold value 870, a white pixel is controlled to be applied to the corresponding position of the input video. Similarly, the comparator 84
The outputs of 0 and 860 are provided to OR gate 920, whose output controls switch 930. That is, the detailed value dv
When either dh or dh is smaller than the negative threshold, the black pixel is controlled to be inserted in the corresponding position of the input video.

The output of switch 930 is provided as output video (eg, to block match and comparator 670).

In another embodiment, the switches 910 and 93 are used.
The replacement of white and black video by 0 is also subject to intensity averaging in small blocks around the current pixel. That is,
Fine areas are replaced with white pixels in areas of low strength, and fine areas are replaced with black pixels in areas of high strength.

In still another specific configuration, a constant signal may be added to the pixel value of the fine area instead of putting black or white pixels in the fine area.

Since the performance of the various implementations described above depends on the video material used, a dynamic system can be used that can switch between these various alternatives.

FIG. 14 illustrates a video enhancement test block 950,9 for a car video (as shown in FIGS. 1 and 2).
The effect on the vector selection process using 60 is shown. FIG.
In FIG. 14, the image enhancement performed by the device of FIG. 14 is shown in bold print to represent the car radio antenna. As a result of image enhancement,
The antenna is a much more important part of the test blocks 950, 960 used for motion vector selection. This means that the correlation test in vector selection is more likely to select the correct vector (the "car" motion vector) to use in interpolating the pixels that represent the car radio antenna.

In another specific configuration of the present invention, the test block used for vector selection is analyzed to detect the amount of detail included in the block. The correlation test results are then adjusted or weighted so that the high degree of correlation obtained with test blocks containing few details (eg, blocks from simple background areas) is reduced to the higher correlation obtained with more detailed test blocks. Treat as less important than the degree of correlation.

The detail level test in the test block is
It can be done in various ways. For example, the dynamic range (maximum pixel value-minimum pixel value) in the test block is quantified. As an improvement, by smoothing the blocks before testing (using a conventional two-dimensional low-pass filter), the otherwise simple 1
One false pixel can be prevented from being received as an indication of detail in the test block. Another one
In one embodiment, Sobel filtering can be used.

Another way to determine the detail value in a block is to calculate the difference between adjacent pixel values and sum those differences for that block.

[0096]

[Equation 1] However, Pixel diff = absolute difference between horizontally adjacent pixels Line diff = absolute difference between vertically adjacent pixels If there is a detail in the block, the difference in luminance value between adjacent pixels will make the above total amount non-zero by an amount according to the high frequency content.

FIG. 15 is a block diagram showing a part of the vector selector of FIG. In the figure, the SAD value generated by the block matching unit 672 is weighted according to the level of detail detected in the test block (1000, 1
010). This means that the results of correlation tests are not rejected simply because the blocks lack detail. Remembering that the high degree of correlation between the test blocks is indicated by the low SAD values generated from the test blocks, the weighting could be done as follows. Difference value = original SAD × weighted value However, when the detail level test result is larger than a predetermined threshold value,
Weight value = threshold / (detail level test result + 1), and if the detail test result is less than or equal to the predetermined threshold value, weight value = 1.

In FIG. 15, the two test blocks (t.b.) pointed to by the motion vector under test.
Of each of the processing units 600, 610, 620, 6
Detail detector 1000 and block matcher 6 in 30
72 in parallel. The SAD value generated by block matcher 672 is provided to multiplier 1010, where the SAD value is multiplied by the weight value generated by detail detector 1000 as described above.

In another specific configuration shown in FIG. 16, the test result indicating the level of detail in the test block is compared with the threshold value. If the level of detail is lower than the threshold value, the correlation test performed with those blocks is ignored.

In the apparatus shown in FIG. 16, RAMs 650, 6
The test block output from 60 is also supplied in parallel to the block matcher 672 and the detail detector 1020. The detail level detector 1020 outputs a detail value indicating the detail level in the test block. This detail value is sent to a comparator 1030 where it is compared to a detail threshold 1040.

The block matcher 672 produces the SAD value from the test block, which is provided to the comparator 1050 along with the current lowest SAD value 675 from the preprocessing unit (see FIG. 7). Comparator 1050 produces an output that indicates whether the SAD value produced by block matcher 672 is less than the current lowest SAD value 675.

The outputs of the comparators 1030 and 1050 are A
It is supplied to the ND gate 1060. The output of the AND gate controls a multiplexer (parallel-to-serial converter) 1070 to produce an output that identifies either the motion vector identifier under test 1080 or the current vector identifier 680 received from a prior processing unit. AND gate 10
60 and multiplexer 1070 operate so that the output of the multiplexer identifies the current motion vector under test only when the following conditions are met: That is, the SAD value for the current tested motion vector is lower than the current lowest SAD value 675 received from the previous processing unit, and the detail detector 1020 in the test block pointed to by the current tested motion vector.
Is greater than or equal to the detail threshold 1040.

Various different embodiments of the present invention have been described above. The characteristics of each of these embodiments (for example, weighting of the zero vector SAD, use of image enhancement, and use of the detail level test) are rearranged in various ways. Note that they may be combined together.

[0104]

As described above, according to the present invention,
When selecting a vector in motion compensation video signal processing,
This has the effect of preventing an incorrect motion vector from being selected due to the apparent correlation between noise regions in the input image.

[Brief description of drawings]

FIG. 1 is a diagram (part 1) showing a problem in a vector selection process for an image of an automobile.

FIG. 2 is a diagram (part 2) showing a problem in a vector selection process for an automobile image.

FIG. 3 is a diagram showing a problem in vector selection processing for an image including a large simple area.

FIG. 4 is a block diagram showing a motion compensation television standard conversion device in which the present invention can be used.

FIG. 5 is an explanatory diagram showing a simple example of vertical space subsampling.

FIG. 6 is an explanatory diagram showing comparison of test blocks when selecting a vector.

FIG. 7 is a block diagram showing a specific example of the vector selector 430 of FIG.

8 is a block diagram showing details of the block matching and comparator 670 of FIG. 7. FIG.

FIG. 9 is a diagram showing a first example of coefficients of a 3 × 3 high-pass spatial filter.

FIG. 10 is a diagram showing a second example of coefficients of a 3 × 3 high-pass spatial filter.

FIG. 11 is a diagram showing coefficients of a vertical Sobel spatial filter.

FIG. 12 is a diagram showing coefficients of a horizontal Sobel spatial filter.

FIG. 13 is a block diagram showing a specific example of a video enhancement circuit.

FIG. 14 is a diagram showing an effect on a vector selection process using a video enhancement test block for the video of the automobile of FIGS. 1 and 2;

FIG. 15 is a block diagram showing another example of the block matching and comparator shown in FIG.

16 is a block diagram showing a modified example of the circuit of FIG.

[Explanation of symbols]

410, 420 Motion vector generating means 430 Vector selector (testing means) 672 Block matching device (correlation value detecting means) 750, 1010 Multiplier (weighting means) 674 Comparator (motion vector selecting means) 340 Motion correction interpolator 810 Vertical Sobel video filter 820 Horizontal Sobel video filter 830, 840, 850, 860 Filter output detection means 910, 930 Pixel replacement means 1000, 1020 Level of detail detector

 ─────────────────────────────────────────────────── ─── Continued Front Page (72) Inventor Stephen Mark Keating UK, Berkshire, Reading, Lower Early, Hunting Don Cloth 28 (72) Inventor Martin Rex Dricot England, Hampshire, Basingstoke, Basing, Ringfield Cloth 6 (72) Cima Rabiji Versani England Hampshire, Basingstoke, Hook, Foracre Coppis 5 (72) Inventor Morgan William Amos David Sally, Farnham, Bloomleaf Road 18 (72) Inventor Kenneth Knight Ramsey, Bergen County, New Jersey, United States ， North Central Avenue 186

Claims (32)

[Claims] Claim: What is claimed is: 1. A means for generating a plurality of motion vectors representing a motion of a video between input video pairs of an input video signal, and a group of motion vectors including a zero motion vector indicating a motion of a zero video. Means for testing the group to select one motion vector to be used for interpolation of the output pixels, the means for testing the correlation degree between the test blocks of the input video pointed to by the tested motion vector. Means for detecting the correlation value shown, and the correlation value indicated by the correlation value for the zero motion vector is higher than the correlation values indicated by the correlation values for the other motion vectors in the group. The means for weighting and the motion vector having the weighted correlation value showing the highest degree of correlation are selected from the above motion vector group. Motion compensated video signal processing apparatus having a means for. 2. The detecting means includes means for detecting an absolute luminance difference between pixel pairs at corresponding positions in a test block, and means for summing the absolute luminance differences to generate the correlation value. The device of claim 1. 3. The apparatus of claim 2 wherein said weighting means includes means for multiplying the correlation value for said zero motion vector by a predetermined constant value less than one. 4. The apparatus according to claim 2, wherein the weighting means includes means for multiplying a correlation value for one or more motion vectors other than the zero motion vector in the group by a weight value larger than 1, respectively. 5. The apparatus of claim 2 wherein said weighting means includes means for subtracting a predetermined constant value from the correlation value for said zero motion vector. 6. The weighting means includes means for adding a weighting value to each correlation value for one or more motion vectors other than the zero motion vector in the group.
Or the device of 5. 7. The weighting means performs an operation of weighting the correlation value such that the degree of correlation represented by the correlation value for each motion vector in the group is reduced by an amount proportional to the magnitude of the motion vector. Claims 1, 2,
The apparatus according to any one of 4 and 6. 8. A step of generating a plurality of motion vectors representing motion of a video between input video pairs of an input video signal, and a group of motion vectors including a zero motion vector indicating motion of a zero video, Testing the group to select one motion vector to be used for interpolation of the output pixels, the testing step comprising determining the correlation between the test blocks of the input video pointed to by the tested motion vector. Detecting the correlation value shown, weighting the correlation value such that the correlation degree indicated by the correlation value for the zero motion vector is greater than the correlation degree indicated by the correlation values for the other motion vectors in the group From the above motion vector group, the motion vector having the weighted correlation value showing the highest correlation is calculated. Motion compensated video signal processing method comprising the step of to-option. 9. A means for generating a plurality of motion vectors representing a motion of a video between input video pairs of an input video signal, and selecting one motion vector used for interpolation of output pixels of an output video from a group of motion vectors. In order to test the group, and a motion compensation interpolator for interpolating output pixels from the input video pair of the input video signal using the selected motion vector. Video enhancement means for enhancing at least the video content of the test block of the input video pointed to by the vector, means for detecting the degree of correlation between the video enhanced test blocks, and A motion compensation video signal processing device having means for selecting a motion vector having a correlation degree. 10. The apparatus of claim 9, wherein the image enhancement means comprises a two-dimensional high pass spatial filter. 11. The image enhancement means comprises: a vertical Sobel video filter; a horizontal Sobel video filter; and a predetermined detail around one or more pixels in a test block, the vertical Sobel video filter or the horizontal Sobel video filter. 10. The apparatus of claim 9 including means for detecting an output indicative of greater than a degree and means responsive to said detecting means for replacing one or more pixels of the test block with pixels of a predetermined intensity. 12. The apparatus of claim 11, wherein the predetermined intensity pixels are white pixels. 13. The image enhancing means detects a second output from the vertical Sobel image filter or the horizontal Sobel image filter, the output indicating less than a predetermined level of detail around one or more pixels in a test block. 13. An apparatus according to claim 11 or 12, comprising detection means and means responsive to the second detection means for replacing one or more pixels of a test block with a second predetermined intensity pixel. 14. The apparatus of claim 13, wherein the second predetermined intensity pixel is a black pixel. 15. The image enhancing means comprises: a vertical Sobel image filter, a horizontal Sobel image filter, and a predetermined degree of detail around one or more pixels in a test block by the vertical Sobel image filter or the horizontal Sobel image filter. 10. The apparatus of claim 9 including means for detecting an output that is greater than, and means responsive to the detecting means for adding a predetermined luminance component to the luminance components of one or more pixels of the test block. 16. A correlation value representing a degree of correlation between video enhancement test blocks of an input video pointed to by a motion vector under test, wherein the group of motion vectors includes a zero motion vector indicating a motion of zero video. Means for weighting the correlation value such that the degree of correlation indicated by the correlation value for the zero motion vector is greater than the degree of correlation indicated by correlation values for the other motion vectors in the group. 16. The apparatus according to claim 9, wherein the selecting means operates to select a motion vector having a weighted correlation value indicating the highest degree of correlation. 17. The detecting means detects the absolute luminance difference between a pair of pixels at corresponding positions in the video enhancement test block, and the absolute luminance difference is summed to generate a sum of absolute differences (SAD) value. A device according to any one of claims 9 to 16 having means for performing. 18. A step of generating a plurality of motion vectors representing a motion of a video between an input video pair of an input video signal, and selecting a motion vector used to interpolate an output pixel of an output video from a group of motion vectors. , Testing the group and interpolating output pixels from an input video pair of an input video signal using a selected motion vector, the testing step being pointed to by a motion vector under test Video enhancement of at least the video content of the test block of the input video, detection of the correlation between the video enhancement test blocks, and selection of the motion vector with the highest correlation between the respective video enhancement test blocks. Method of motion compensation video signal processing including steps. 19. Means for generating a plurality of motion vectors representing the motion of a video between input video pairs of an input video signal, and a group of motion vectors for selecting a motion vector used to interpolate output pixels of an output video. And means for detecting the correlation between the test blocks of the input video pointed to by the motion vector to be tested, and generating an analysis value indicating the video detail of the test block. And a test block pointed to by the motion vector,
Means for selecting one motion vector from the group of motion vectors according to the degree of correlation with the analysis value for the test block pointed to by the motion vector. 20. The apparatus of claim 19, wherein the means for generating the parsed value includes means for detecting a difference between a highest intensity component of a pixel in a test block and a lowest intensity component of a pixel in the test block. 21. The apparatus of claim 19, wherein the means for generating the analytic value includes means for summing the luminance differences between adjacent pixel pairs within each test block. 22. The summing means includes, for each pixel in the test block, a luminance difference between a higher luminance difference between the pixel and a vertically adjacent pixel and a luminance between the pixel and a horizontally adjacent pixel. 23. The apparatus of claim 22, including means for adding the difference and. 23. The method according to claim 19, further comprising means for preventing selection of a motion vector having an analysis value indicating that the test block pointed to by the motion vector is lower than a predetermined image detail level. apparatus. 24. A means for detecting a correlation value indicating a correlation degree between test blocks, and a correlation degree indicated by the correlation value for a motion vector, according to an analysis value, increases as the detected image detail level increases. 23. The apparatus according to claim 19, further comprising means for weighting the correlation value, and the selecting means operates to select a motion vector having a weighted correlation value indicating the highest degree of correlation. 25. The means for detecting the degree of correlation includes a means for detecting an absolute luminance difference between pixel pairs at corresponding positions in a test block, and a sum of the absolute luminance differences for summing absolute difference (SAD) values. 25. The apparatus of claim 24, comprising: 26. The apparatus of claim 25, wherein the weighting means includes means for multiplying the total absolute difference value by a weighting factor that decreases as the detected video detail increases. 27. An apparatus according to any one of claims 19 to 26, comprising image enhancing means for enhancing the image content of at least the test block of the input image pointed to by the motion vector under test. 28. Means for detecting a correlation value indicating a degree of correlation between test blocks of an input video pointed to by a motion vector under test, wherein the group of motion vectors includes a zero motion vector indicating a motion of zero video. , Means for weighting the correlation value such that the degree of correlation indicated by the correlation value for the zero motion vector is increased relative to the degree of correlation indicated by correlation values for the other vectors in the group, and The apparatus according to any one of claims 19 to 27, wherein the selecting means operates to select a motion vector having a weighted correlation value indicating the highest degree of correlation. 29. The input video pair comprises two consecutive fields of an interlaced input video signal.
The apparatus of any one of .about.17 or 19-28. 30. The output video contains one field of an interlaced output video signal.
29. The apparatus according to any one of 29 to 29. 31. A step of generating a plurality of motion vectors representing a motion of a video between an input video pair of an input video signal, and a group of motion vectors for selecting a motion vector used to interpolate an output pixel of an output video. And detecting the correlation between the test blocks of the input video pointed to by the motion vector under test, and generating an analysis value indicating the spatial frequency content of the test blocks. , A test block pointed to by a motion vector,
A motion compensated video signal processing method comprising the steps of selecting a motion vector from the group of motion vectors according to a degree of correlation with an analysis value for a test block pointed to by the motion vector. 32. A method according to claims 1 to 7, 9 to 17 or 19 to 1.
A television standard format conversion device comprising the device according to any one of 3 above.