JPH089340A - Motion vector detector and motion vector detection method - Google Patents

Abstract

PURPOSE:To reduce the deterioration in the image quality due to block distortion caused by an isolated vector by discriminating the isolated vector from the detected motion vector and eliminating the isolated vector. CONSTITUTION:A motion vector detection section 2 detects a motion vector of a video signal received from an input terminal 1. A motion vector memory 3 stores tentatively a motion vector required to discriminate an isolated vector among motion vectors detected by the motion vector detection section 2. An isolated vector discrimination section 4 takes correlation with surrounding vectors as to each of detected motion vectors to discriminate whether, or not the object motion vector is an isolated vector. An isolated vector elimination section 5 eliminates a motion vector discriminated to be the isolated vector by the isolated vector discrimination section 4. The motion vector after isolated vector elimination processing is outputted from an output terminal 6.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a motion vector detecting device and a motion vector detecting method necessary for scanning conversion and frame number conversion using motion compensation.

[0002]

2. Description of the Related Art As a method for detecting a motion vector, a block matching method has been well known. FIG. 9 is a block diagram showing an example of the principle configuration of a conventional motion vector detection device disclosed in, for example, Japanese Patent Laid-Open No. 2-52914. In the figure, 28 is a frame memory that holds an input video signal for one frame period, 29 is 2 frames (n-1) at the input and output of the frame memory 28,
A correlation calculation unit (here, (n) indicates a current frame, (n-1) indicates a preceding frame) for calculating a correlation regarding a target block which is a target of motion vector detection between images of (n), and 30. A shift vector generation unit that determines the relative position of the comparison block with respect to the target block, 31 takes out the minimum value from each operation value group obtained by the correlation calculation unit 29, and the minimum gate that determines the motion vector, 32 detects It is a motion vector memory that stores motion vectors.

Next, the operation of the conventional example will be described. First, the input video signal of frame (n-1) is input to the frame memory 28 and held for one frame period. Then, the correlation calculation unit 29 calculates the correlation of the target block for the motion vector detection between the images of the two frames (n-1) and (n) which are the input and the output of the frame memory 28. At that time, the preceding frame (n-
The relative position of the comparison block of 1) is set by the shift vector generation unit 30. At the same time as extracting the maximum correlation value from the respective correlation value groups of the target block of the comparison block by the minimum gate 31 for detecting the minimum value and extracting the motion vector output corresponding to the maximum correlation value. , The motion vector output value is stored in the motion vector memory 32, and is used as a control input to the shift vector generation unit 30 when calculating the correlation similarly for the subsequent frame (n + 1), and the motion vector output is output as described above. The position of the comparison block shifted corresponding to the value is set.

[0004]

The conventional motion vector detecting device as described above can reduce the motion vector detection time and is suitable for use in information compression such as image communication. When used for scanning line interpolation or frame number conversion of a John device etc., when an isolated motion vector (isolated vector) that has no correlation with the surrounding motion vector due to the false detection of the motion vector occurs, it becomes block distortion, There was a problem that the image quality deteriorated significantly.

The present invention has been made in order to solve the above problems, and in the processing such as scanning line conversion and frame number conversion by judging and removing an isolated vector from the detected motion vector, The purpose is to detect a motion vector with little deterioration in image quality.

[0006]

According to a first aspect of the present invention, there is provided a motion vector detecting device for detecting a motion vector from an input video signal, wherein one screen is divided into a plurality of blocks, and each block has a motion vector. A motion vector detecting means for detecting the motion vector, and an isolated vector for determining whether or not the motion vector of the block of interest is an isolated vector by using the motion vector of the block of interest as the result of motion vector detection and the comparison block for the block of interest. When the determination unit and the isolated vector determination unit determine that the motion vector of the block of interest is an isolated vector, a new motion of the block of interest is obtained from the motion vector detection results of the block of interest and the comparison block. And an isolated vector removing means for determining a vector.

According to a second aspect of the present invention, in the isolated vector determination means, the determination of the correlation between the target block and the comparison block for the target block is performed by comparing with the motion vector of the target block which is the motion vector detection result. The absolute value of the difference between the size of each block and the motion vector is used, and when this absolute value exceeds a specified value, it is assumed that the target block and the comparison block have no correlation, and the same calculation is performed on the target block. For all the comparison blocks, the total number of comparison blocks having no correlation with the target block is accumulated, and the isolated vector for the motion vector of the target block is determined based on the accumulated value.

According to a third aspect of the present invention, the isolated vector determining means compares and determines the correlation between the block of interest and the comparison block for the block of interest by comparing with the motion vector of the block of interest that is the motion vector detection result. The difference value in each horizontal component and vertical component with respect to the motion vector of each block is used, and when this difference value exceeds a specified value, there is no correlation between the target block and the comparison block in the horizontal direction or the vertical direction. The same calculation is performed on all the comparison blocks for the target block, the number of comparison blocks having no correlation with the target block is accumulated for each component, and the motion vector of the target block is calculated based on the accumulated value. The determination of an isolated vector is performed.

According to a fourth aspect of the present invention, the isolated vector removing means newly determines the motion vector of the target block which is an isolated vector, by determining each of the motion vectors of the comparison blocks corresponding to the target block which is the motion vector detection result. Based on the horizontal and vertical components, the largest value for each component is set as a changed value, that is, a new vector component of the target block, and when there are a plurality of the same number of component values, the changed value is determined in accordance with the specified priority order. Is to do.

A motion vector detecting method according to a fifth aspect of the present invention is a method for detecting a motion vector from an input video signal, wherein one screen is divided into a plurality of blocks, and the motion vector is detected for each block. It is determined whether the motion vector of the target block is an isolated vector by using the motion vector of the target block that is the detection result and the comparison block for the target block, and the motion vector of the target block is an isolated vector. When it is determined, a new motion vector of the target block is determined using the motion vectors of the target block and the comparison block, which are the motion vector detection results.

According to a sixth aspect of the present invention, in the isolated vector determination method, the determination of the correlation between the target block and the comparison block for the target block is performed by comparing with the motion vector of the target block which is the motion vector detection result. The absolute value of the difference between the size of each block and the motion vector is used, and when this absolute value exceeds a specified value, it is assumed that the target block and the comparison block have no correlation, and the same calculation is performed on the target block. For all the comparison blocks, the total number of comparison blocks having no correlation with the target block is accumulated, and the isolated vector for the motion vector of the target block is determined based on the accumulated value.

According to a seventh aspect of the present invention, in the isolated vector determination method, the correlation between the target block and the comparison block for the target block is compared and determined, and the motion vector of the target block that is the motion vector detection result and the comparison block are compared. A difference value between each horizontal component and a vertical component with respect to a motion vector for each block is used, and when the difference value exceeds a specified value, there is no correlation between the target block and the comparison block in the horizontal direction or the vertical direction. Then, the same calculation is performed on all the comparison blocks for the target block, the number of comparison blocks having no correlation with the target block is accumulated for each component, and the motion vector of the target block is isolated based on the accumulated value. This is to determine a vector.

According to the eighth aspect of the present invention, in the isolated vector removing method, the motion vector of the block of interest that is an isolated vector is newly determined for each of the motion vectors of the comparison block with respect to the block of interest that is the motion vector detection result. Based on the horizontal and vertical components, the largest value for each component is set as a changed value, that is, a new vector component of the target block, and when there are a plurality of the same number of component values, the changed value is determined in accordance with the specified priority order. Is to do.

[0014]

According to the first aspect of the present invention, the motion vector is detected for each block, it is determined whether the motion vector for each block is an isolated vector, and when the isolated vector exists, the isolated vector is removed.

According to a second aspect of the present invention, the operation of obtaining the correlation between the target block and the comparison block for the target block in the isolated vector determining means is performed by the motion vector of the target block which is the motion vector detection result and the comparison block. The isolated vector for the motion vector of the target block is determined by using the absolute value of the difference between the size of the motion vector of each block and the number of comparison blocks that have no correlation with the target block.

According to a third aspect of the present invention, the operation of obtaining the correlation between the target block and the comparison block for the target block in the isolated vector determination means is performed by the motion vector of the target block which is the motion vector detection result and the comparison block. The motion vector of the block of interest considering the direction of the vector is calculated by using the difference value in each horizontal component and vertical component with respect to the motion vector of each block, and obtaining the number of comparison blocks having no correlation with the block of interest. Determine the isolated vector.

According to a fourth aspect of the present invention, the operation of newly determining the motion vector of the target block which is an isolated vector in the isolated vector removing means is performed by the horizontal and vertical components of the motion vector of the comparison block with respect to the target block. On the basis of the above, the maximum value for each component is set as a change value, that is, a new vector component of the target block, and when there are a plurality of the same number of component values, the change value is determined according to the prescribed priority order.

According to the invention of claim 5, the motion vector is detected for each block, it is determined whether the motion vector for each block is an isolated vector, and if the isolated vector exists, the isolated vector is removed. .

According to the sixth aspect of the present invention, the operation of obtaining the correlation between the block of interest and the comparison block for the block of interest in the isolated vector determination method is performed by the motion vector of the block of interest which is the motion vector detection result and the comparison block 1. An isolated vector for the motion vector of the target block is determined by using the absolute value of the difference between the size of the motion vector of each block and the number of comparison blocks having no correlation with the target block.

According to a seventh aspect of the present invention, the operation of obtaining the correlation between the block of interest and the comparison block for the block of interest in the isolated vector determination method is performed by the motion vector of the block of interest, which is a motion vector detection result, and the comparison block 1. Isolation with respect to the motion vector of the target block in consideration of the direction of the vector by calculating the number of comparison blocks having no correlation with the target block by using the difference value in each horizontal component and vertical component with the motion vector of each block Determine the vector.

According to the eighth aspect of the invention, the operation of newly determining the motion vector of the target block which is an isolated vector in the isolated vector removing method is performed by the horizontal and vertical components of the motion vector of the comparison block for the target block. On the basis of the above, the maximum value for each component is set as a change value, that is, a new vector component of the target block, and when there are a plurality of the same number of component values, the change value is determined according to the prescribed priority order.

[0022]

【Example】

Example 1. Before describing the embodiments of the present invention, the principle of scanning line conversion using motion vectors will be described. Figure 10
As an example of the scanning line conversion using the motion vector, the case where the interlaced signal which is the original signal is sequentially scanned and converted is shown. Here, it is assumed that the current field is the second field 34 and scanning line interpolation is performed. First, the direction and distance (motion vector) that the subject has moved between frames are calculated. As a result, the position of the subject in the immediately preceding first field 33 is known. Based on this, the scanning line interpolation is performed using the pixels shifted by the amount of the motion vector between the fields, so that the progressive scan conversion image 36 having the vertical resolution equivalent to the original signal can be obtained.

However, when the first field 33 and the second field 34 are overlapped without using the motion vector, the progressive scan conversion image 35 because the positions of the images of the first field 33 and the second field 34 are different. As described above, the image has a reduced vertical resolution, and when a motion vector is used, it becomes an image 36. The motion vector detecting device according to the present invention is used for the above-mentioned applications such as scanning line interpolation and frame conversion.

Next, a first embodiment of the present invention will be described.
FIG. 1 is a block diagram showing a motion vector detection device of the first embodiment. In FIG. 1, 1 is an input terminal to which a video signal is input, 2 is a motion vector detection unit that detects a motion vector of the input video signal, and 3 is a motion vector detection unit 2.
Among the motion vectors detected by, the motion vector memory 4 for temporarily storing motion vectors necessary for the determination of an isolated vector, which will be described later, is correlated with the surroundings for each detected motion vector, and An isolated vector determination unit that determines whether a certain motion vector is an isolated vector, 5 is an isolated vector removal unit that removes a motion vector that has been determined by the isolated vector determination unit 4 to be an isolated vector, and 6 is an isolated vector removal processing This is an output terminal for outputting a motion vector.

FIG. 2 shows a block circuit diagram of the motion vector detecting section 1 of FIG. In FIG. 2, an interlaced signal is used as the video input signal. In FIG. 2, an input terminal 1 is input with an interlace signal in which one screen is composed of two fields.
2 field period, that is, one screen is held by 8,
Then, the correlation calculation unit 9 calculates the correlation between the image of the preceding field n−1, which is the input of the field memory 7, and the subsequent field n + 1, which is the output of the field memory 8, regarding the target block for the motion vector detection. . Such correlation calculation is performed for all comparison blocks that are comparison targets of the target block, the motion vector extraction unit 10 extracts the motion vector of the comparison block having the maximum correlation, and outputs it from the output terminal 11.

FIG. 3 shows a block circuit diagram of the isolated vector determination section 4 of FIG. In FIG. 3, two input terminals 1
Of the motion vectors obtained by the motion vector detecting unit 2, reference numerals 2 and 13 respectively indicate the motion vector of the target block,
A motion vector group corresponding to each comparison block for obtaining the correlation with the target block is input. Here, the following description will be given assuming that the motion vector corresponding to the target block is A and the motion vector corresponding to the comparison block is B.

First, the correlation calculator 14 obtains the correlation between the input motion vectors A and B. This is done by taking the absolute value of the difference between the magnitudes of the motion vectors A and B. The calculated value is compared with a specified value in the accumulator 15, and if it exceeds the specified value, it is considered that there is no correlation between the motion vectors of the target block and the comparison block, and the cumulative number S is counted. Hereinafter, a similar comparison operation is performed on comparison blocks around the target block, for example, 8 blocks near the target block as shown in FIG. 4, to obtain a cumulative value S. Next, the determination unit 16 compares the cumulative value S obtained above with a specified value. When S exceeds the specified value, it is determined that the motion vector corresponding to the target block is an isolated vector, and the isolated vector determination flag is set. L is set to L = 1, and when it is not an isolated vector, it is set to L = 0 and output from the output terminal 17.

FIG. 5 shows a block circuit diagram of the isolated vector removing unit 5 in FIG. In FIG. 5, as the three input values, when the motion vector A of the target block obtained above and the motion vector of the target block are isolated vectors, the blocks around the target vector necessary for the removal process of the isolated vector The motion vector group B corresponding to a certain comparison block and the isolated vector determination flag L obtained by the isolated vector determination unit 4 are input. Further, in this embodiment,
The isolated vector determination flag L determines whether or not the isolated vector is removed. That is, when L = 0, the isolated vector is not removed, the motion vector corresponding to the target block is output, and when L = 1, the isolated vector removal processing is performed to change the motion vector corresponding to the target block. Then, the changed value is output from the output terminal 24.

The operation of the isolated vector removing section 5 will be described below. In FIG. 5, with respect to the input motion vector group B of the comparison block for the target block obtained above, statistics are obtained for each of the horizontal and vertical components by the integrating units 18 and 20. Next, the component changing units 19 and 21 extract the most horizontal and vertical components as change value candidates, and then select the change value from the change value candidates according to the priority order shown in FIG. The signal is output from the output terminal 6 via the selector 22 controlled by L.

Here, how a change value is selected from the change value candidates will be described below. It is assumed that eight comparison blocks have the component values X, Y, and Z as shown in FIG. 7 for the block of interest. In this case, the component value X and the component value Y exist in the same number, three each, and become the change value candidates. Therefore, in accordance with the priority order shown in FIG. 6, the highest priority order of each change value candidate is obtained and compared, and the change value candidate having the higher priority order is selected. In the case of FIG. 7, the highest priority of the change value candidate X is 5 and the highest priority of the change value candidate Y is 2, so that Y is the change value to be obtained.

By the way, in the above description, an example in which the present invention is realized by hardware is shown, but the present invention may be realized by software.

The operation of the present invention when implemented by software will be described below. First, for an interlaced signal in which one screen, which is a video input signal, is composed of two fields, one screen is divided into a plurality of blocks, and the preceding field n-1 with respect to the current field n is divided for each block.
And a motion vector between the images of the following field n + 1.

Next, it is determined whether or not the motion vector of the target block is an isolated vector by using the correlation value between the motion vector of the target block and the motion vector of the comparison block for the target block. For this correlation value, the absolute value of the difference in the magnitude of the motion vector between the block of interest and the comparison block is used. If this absolute value exceeds a specified value, there is a gap between the block of interest and the comparison block. Considered uncorrelated. The same calculation is performed on all the comparison blocks for the target block to obtain the number of comparison blocks having no correlation with the target block. When the number exceeds a specified value, the motion vector of the target block is an isolated vector. It is determined that

Next, as the isolated vector removal processing, the component value of the motion vector determined to be an isolated vector is changed. First, the statistics of the horizontal and vertical components of the motion vector of the target block and all the comparison blocks for the target block are obtained for each component, and the statistical values show the most horizontal and vertical components of the motion vector of the target block. The change value of each component is selected as a candidate, and the change value is further selected according to the priority order shown in FIG.

Example 2. FIG. 8 is a block circuit diagram of an isolated vector determination unit according to the second embodiment of the present invention. Since the other components are the same as those of the first embodiment of FIG.
Only the operation of the isolated vector determination unit 5, which is a different part from the above, will be described. In FIG. 8, input terminals 12 and 13 are supplied with motion vectors A and B similar to those in the first embodiment. Here, the horizontal and vertical components of the motion vectors A and B are A = (Ax, Ay) and B = (Bx, By), respectively, and the difference value of each component is Cx = Ax−Bx, Cy =, respectively. Ay-
It is represented by By.

The operation of this embodiment will be described below. First, when obtaining the correlation between the input motion vectors A and B,
Difference values Cx and Cy for each component are used as correlation values in the correlation calculation unit 25.
Ask for. These Cx and Cy are compared with the prescribed values for each direction in the accumulator 26, and if they are out of the regulation, it is considered that there is no correlation between the motion vectors A and B, and the cumulative variable S is counted. Hereinafter, a similar comparison operation is performed on comparison blocks around the target block, for example, eight blocks in the vicinity of the target block as shown in FIG.
Ask for. The operation of the determination unit 16 and thereafter is similar to that of the first embodiment, and the processed motion vector determination flag L is output to the output terminal 17.
Will be output.

By the way, in the above description, an example in which the present invention is realized by hardware is shown, but the present invention may be realized by software.

The operation of the present invention when implemented on software will be described below. As described above, the motion vector of the target block is A = (Ax, Ay), and the motion vector of the comparison block for the target block is B = (Bx, Ay).
By) and the difference value of each component is Cx = Ax
-Bx, Cy = Ay-By. First, the correlation value between the motion vector A of the target block and the motion vector B of the comparison block for the target block is obtained. In this correlation value,
The difference values Cx and Cy for each component are used, and each is compared with a specified value for each component, and if the specified value is exceeded, it is considered that there is no correlation between the target block and the comparison block in the horizontal or vertical direction. , The same operation is performed for all comparison blocks for the target block,
Calculate the number of comparison blocks that have no correlation with the block of interest,
When the number exceeds the specified value, it is determined that the motion vector of the target block is an isolated vector.

By the way, in the description of the first and second embodiments,
The case of an interlaced signal that composes one screen in two fields as an input video signal has been described, but in addition to this, even when a non-interlaced signal, an image from a movie film, etc. is input, one frame or one frame constitutes one screen. The present invention can be applied in the same manner by considering that the above has been done.

In the description of the first and second embodiments, the neighborhood 8 shown in FIG. 4 is used as the comparison block for the target block.
Although the case of using one block has been described, the number of comparison blocks may be increased or decreased from eight blocks.
Further, in the description of the above-described first and second embodiments, the case of FIG. 6 is described as the priority order of the comparison blocks, but the priority order may be changed.

[0041]

As described above, according to the first aspect of the invention, in the motion vector detecting device, the motion vectors of the target block and the comparison block for the target block are calculated for all the calculated motion vectors. It is possible to remove the isolated vector by performing an isolated vector determination using, and newly determining a motion vector for the motion vector of the block of interest determined to be the isolated vector by the isolated vector determination. Therefore, in the processing such as the scanning line conversion and the frame number conversion, it is possible to reduce the image quality deterioration due to the block distortion caused by the isolated vector.

According to the second aspect of the present invention, in the isolated vector determining means, the difference in magnitude between the motion vector of the target block which is the motion vector detection result and the motion vector of each comparison block for the target block. By determining the correlation between the target block and the comparison block using the absolute value of, and further determining the number of comparison blocks having no correlation with the target block, it is possible to easily determine the isolated vector for the motion vector of the target block. It can be performed.

Further, according to the third aspect of the invention, in the isolated vector determination means, each horizontal component of the motion vector of the target block which is the motion vector detection result and the motion vector of each comparison block for the target block, By using the difference value in the vertical component, the correlation between the target block and the comparison block is calculated, and the number of comparison blocks having no correlation with the target block is calculated to obtain the motion vector of the target block in consideration of the vector direction. It is possible to determine the isolated vector for.

According to the fourth aspect of the invention, when the motion vector of the target block is an isolated vector, the isolated vector removing means determines, based on each horizontal component and vertical component of the motion vector of the comparison block for the target block,
The largest value for each component is the changed value, that is, the new vector component of the block of interest, and when there are a plurality of the same number of component values, the changed value is determined according to the specified priority order, and the correlation with the surroundings It is possible to change the motion vector in consideration of the above, and thus the isolated vector can be removed.

According to the fifth aspect of the invention, in the motion vector detecting device, for all the obtained motion vectors, the motion vector of the target block and the comparison block for the target block is used to isolate the motion vector. By performing the vector determination and newly determining the motion vector for the motion vector determined as the isolated vector, it becomes possible to remove the isolated vector. Therefore, processing such as scanning line conversion and frame number conversion is performed. In, the image quality deterioration due to block distortion caused by the isolated vector can be reduced.

According to the sixth aspect of the invention, in the isolated vector determination method, the difference in magnitude between the motion vector of the target block which is the motion vector detection result and the motion vector of each comparison block for the target block. By determining the correlation between the target block and the comparison block using the absolute value of, and further determining the number of comparison blocks having no correlation with the target block, it is possible to easily determine the isolated vector for the motion vector of the target block. It can be performed.

Further, according to the invention of claim 7, in the isolated vector determination method, each horizontal component of the motion vector of the target block which is the motion vector detection result and the motion vector of each comparison block for the target block, By using the difference value in the vertical component, the correlation between the target block and the comparison block is calculated, and the number of comparison blocks having no correlation with the target block is calculated to obtain the motion vector of the target block in consideration of the vector direction. It is possible to determine the isolated vector for.

According to the invention of claim 8, when the motion vector of the target block is an isolated vector, in the isolated vector removing method, based on each horizontal component and vertical component of the motion vector of the comparison block for the target block,
The largest value for each component is the changed value, that is, the new vector component of the block of interest, and when there are a plurality of the same number of component values, the changed value is determined according to the specified priority order, and the correlation with the surroundings It is possible to change the motion vector of the block of interest in consideration of the above, and thus the isolated vector can be removed.

[Brief description of drawings]

FIG. 1 is a block diagram showing a motion vector detection device according to a first embodiment of the present invention.

FIG. 2 is a block circuit diagram showing a motion vector detection unit of the first embodiment.

FIG. 3 is a block circuit diagram showing an isolated vector determination unit according to the first embodiment.

FIG. 4 is a diagram showing an example of a position of a comparison block with respect to a target block.

FIG. 5 is a block circuit diagram showing an isolated vector removing unit according to the first embodiment.

FIG. 6 is a diagram showing a priority order when changing a component of a target block in the isolated vector removing unit of the first embodiment.

FIG. 7 is a diagram showing an example of values taken by a comparison block for a target block.

FIG. 8 is a block circuit diagram showing an isolated vector determination unit according to the second embodiment of the present invention.

FIG. 9 is a diagram showing a conventional motion vector detection device.

FIG. 10 is a diagram showing the principle of scanning line conversion using motion vectors.

[Explanation of symbols]

1 input terminal, 2 motion vector detection section, 3 motion vector memory, 4 isolated vector determination section, 5 isolated vector removal section, 6 output terminals, 7 and 8 field memory,
9 correlation calculation unit, 10 motion vector extraction unit, 11 output terminal, 12 input terminal, 13 input terminal, 14 correlation calculation unit, 15 integration unit, 16 determination unit, 17 output terminal, 18, 20 integration unit, 19, 21 change Part, 22
Input terminal, 23 selector, 24 output terminal, 25 correlation calculation unit, 26 integration unit, 27 output terminal, 28 frame memory, 29 correlation calculation unit, 30 shift vector generator, 31 minimum gate, 32 motion vector memory, 33, 34 Interlaced signal, 35, 36 progressive scan conversion image.

Front page continuation (72) Inventor Takuji Kurashita No. 1 Baba Institute, Nagaokakyo-shi Video System Development Laboratory, Mitsubishi Electric Corporation

Claims (8)

[Claims] 1. A device for detecting a motion vector from an input video signal, wherein one screen is divided into a plurality of blocks, and a motion vector detection means for detecting a motion vector for each block, and a motion vector detection result. An isolated vector determining means for determining whether or not the motion vector of the target block is an isolated vector using the motion vector of the block and the comparison block for the target block, and the motion vector of the target block by the isolated vector determining means. If is determined to be an isolated vector,
A motion vector detecting device, comprising: an isolated vector removing unit that determines a new motion vector of the target block from the motion vector detection results of the target block and the comparison block. 2. The isolated vector determination means compares and determines the correlation between the block of interest and the comparison block for the block of interest for each motion block of the block of interest, which is a motion vector detection result, and for each block of the comparison block. The absolute value of the magnitude difference from the motion vector is taken, and when this absolute value exceeds a specified value, it is assumed that there is no correlation between the target block and the comparison block, and the same calculation is performed for all comparison blocks for the target block. The total number of comparison blocks that have no correlation with the target block,
The motion vector detecting device according to claim 1, wherein the motion vector detecting device determines the isolated vector for the motion vector of the block of interest based on the cumulative value. 3. The isolated vector determination means compares and determines the correlation between the target block and the comparison block for the target block for each motion block of the target block, which is the motion vector detection result, and for each one block of the comparison block. The difference value between each horizontal component and the vertical component with respect to the motion vector is taken, and when this difference value exceeds a specified value, it is assumed that there is no correlation between the block of interest and the comparison block in the horizontal direction or the vertical direction. The calculation is performed on all the comparison blocks for the target block, the number of comparison blocks that do not correlate with the target block is accumulated for each component, and the isolated vector for the motion vector of the target block is determined based on the accumulated value. The motion vector detection device according to claim 1, wherein the motion vector detection device is performed. 4. The isolated vector removing means newly determines the motion vector of the target block that is an isolated vector by determining each horizontal component and vertical component of the motion vector of the comparison block with respect to the target block that is the motion vector detection result. On the basis of the above, the maximum value for each component is used as a change value, that is, a new vector component of the target block, and when there are a plurality of component values of the same number, the change value is determined according to a prescribed priority order. The motion vector detection device according to claim 1. 5. A method for detecting a motion vector from an input video signal, wherein one screen is divided into a plurality of blocks, a motion vector is detected for each block, and a target block which is a motion vector detection result and the target block. The motion vector of the target block is determined by using the motion vector with the comparison block for the target block, and the motion vector detection result is determined when the motion vector of the target block is determined to be an isolated vector. The motion vector detecting method is characterized in that a new motion vector of the target block is determined by using motion vectors of the target block and the comparison block. 6. In the isolated vector determination method, the correlation between a block of interest and a comparison block for the block of interest is compared and determined for each block of the motion vector of the block of interest, which is a motion vector detection result, and each block of the comparison block. The absolute value of the magnitude difference from the motion vector is taken, and when this absolute value exceeds a specified value, it is assumed that there is no correlation between the target block and the comparison block, and the same calculation is performed for all comparison blocks for the target block. The total number of comparison blocks that have no correlation with the target block,
The motion vector detecting method according to claim 5, wherein an isolated vector for the motion vector of the target block is determined based on the cumulative value. 7. In the isolated vector determination method, the correlation between a block of interest and a comparison block for the block of interest is compared and determined by the motion vector of the block of interest, which is a motion vector detection result, and the motion of each block of the comparison block. A difference value in each horizontal component and vertical component from the vector is taken, and when this difference value exceeds a specified value, it is assumed that there is no correlation between the target block and the comparison block in the horizontal direction or the vertical direction, and the same calculation is performed. Is performed for all the comparison blocks for the target block, the number of comparison blocks that do not correlate with the target block for each component is accumulated, and the isolated vector for the motion vector of the target block is determined based on the accumulated value. The motion vector detecting method according to claim 5, wherein 8. In the isolated vector removing method, the motion vector of the target block that is an isolated vector is newly determined by determining each horizontal component and vertical component of the motion vector of the comparison block with respect to the target block that is the motion vector detection result. On the basis of the above, the maximum value for each component is used as a change value, that is, a new vector component of the target block, and when there are a plurality of component values of the same number, the change value is determined according to a prescribed priority order. The motion vector detecting method according to claim 5.