JPH08237660A - Picture processor and mthod therefor - Google Patents

Abstract

PURPOSE: To make the scale of the entire device small in a motion detection circuit using so-called block matching. CONSTITUTION: In an adder 8, the coordinate 39 of a reference block, a motion vector 44 on a circuit operation and the offset 46 of an original point are added and the coordinate (Xc, Yc) of a candidate block is calculated. In a comparator 13, based on the residual 12 of the picture data S2 of the reference block of a present screen and the picture data 51 of the search range of a previous screen and the (Xc, Yc) the validity of the motion vector 44 on the circuit operation is judged and the motion vector 44 on the circuit operation most suitable for a picture compression processing is decided. The motion vector 44 on the circuit operation is stored in a register 15 and thereafter, outputted through an output terminal 19.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus and method for detecting a motion vector used for compression processing of a moving image by using a so-called block matching method.

[0002]

2. Description of the Related Art Motion detection is an important technique used for data compression of moving images. For example, consider transmitting a series of images as shown in FIG. 8 in order. The previous screen 51 shown in FIG. 8 is an image already transmitted, and the current screen 50 is an image to be transmitted. The amount of moving image data can be reduced by paying attention to the local movement of the image. For example, as shown in FIG.
The content of the block 52 of 0 is the block 53 of the previous screen 51.
It is assumed that the contents of can be substituted. Then, instead of transmitting the content of the block 52 as it is, by transmitting the data of the motion vector 57 shown in FIG. 9 to show the positional relationship between the block 52 and the block 53, a significant data compression can be expected. .

As a method (motion detection) for actually obtaining a motion vector, "block matching" is generally used.
Is used. The principle of block matching will be described in the following example. First, terms are defined using FIG. As shown in FIG. 10, the current screen 50 is divided into blocks in advance. Each of the divided blocks is used as a reference block 54. On the other hand, the search range 55 on the previous screen 51 is a fixed area centered on the position of the reference block 54. Further, the candidate block 56 is an arbitrary block included in the search range 55. In addition, the candidate block 56 viewed from the reference block 54
Is referred to as a motion vector 57. Motion vector 5
The unit of 7 is the number of pixels.

FIG. 11A shows a pixel pattern forming the reference block 54 when the size of the reference block 54 is 4 × 4 pixels, and FIG. 11B shows a search range 55 of 8 × 8 pixels. It is a figure which shows the pixel pattern which comprises the search range 55 in the case of. The value of each pixel (generally the brightness value is used) is Pi, j (i = 2,3,4,5; j = 2,3,4,5) for the reference block 54, and the search range is 5
For No. 5, Qi, j (i = 0,1, .., 7; j = 0,1, .., 7) was set. Therefore, the motion vector origin 58, that is, the center position of the reference block 54, coincides with the center position of the search range 55.

Next, a candidate block 56 is selected from the search range 55. The position of the candidate block 56 is determined by the value of the motion vector 57. For example, in the candidate block 56 shown in FIG. 12, the value of the motion vector 57 is (−2, −).
It is for 2). Also, FIG. 13, FIG. 14, and FIG.
5, the motion vector 57 is (-2, -1),
The candidate block 56 at the time of (-2, +2), (+2, +2) is shown.

In this way, the value of the motion vector 57 is 5 for the X and Y components, from -2 to +2, respectively.
Since there are various possible values, a total of 25 candidate blocks 56 can be selected. The 25 selected candidate blocks 56 are respectively compared with the reference block 54. Specifically, the motion vector (x, y) (x = -2,-
1, .., 2; y = -2, -1, .., 2), the residual Dx, y defined by the following equation (1) is obtained.

[0007]

[Equation 1]

As can be seen from the above equation (1), the value of the residual Dx, y becomes zero only when all the pixel values of the candidate block 56 and the reference block 54 match. Therefore, the value of the residual Dx, y can be considered as an index indicating the degree of approximation between the candidate block 56 and the reference block 54. Therefore, of all the motion vectors 57, the one that minimizes the value of Dx, y is selected, and the optimum motion vector 57 is selected.
Decide. That is, if Dx0, y0 is defined by the following equation (2),

[0009]

[Equation 2]

(X0, y0) is the optimum motion vector 57. The motion vector finally output from the motion detection circuit is the value of (x0, y0). The above is a specific description of motion detection by block matching.

FIG. 16 shows the reference block 54 and the search range 5.
FIG. 5 is a diagram showing a general positional relationship of a motion vector detection range 60. By the way, with respect to the reference blocks located at the top, bottom, left, and right ends of the screen, there is no valid screen data outside, so that the value that the motion vector can take is limited. FIG. 10 is a diagram for explaining the reference block located at the left end of the screen. As shown in FIG. 10, in the reference block 54 located at the left end of the screen, the X-direction component of the motion vector 57 is limited to zero or a positive value.

In terms of circuit operation, since the invalid pixel data 61 is input to a part of the search range 55 as shown in FIG. 18, the motion vector range 60 must be limited to the range shown by the dotted line. Therefore, in the conventional motion detection circuit, a 4-bit input terminal called "edge control" is provided and a control signal from the outside is added thereto to realize the above-described limiting function. In particular,
When the edge control terminals are EC0 to EC3, the following control is performed.

First, when the signal "0" is input to all of EC0 to EC3, the value of the motion vector is not limited. On the other hand, when the signal "1" is input to any of the terminals, the following operation is performed according to the input terminal.

EC0: an operation mode in which a reference block at the top of the screen is handled. A motion vector whose Y component is negative is not output. EC1: An operation mode that handles the reference block at the right end of the screen. A motion vector whose X component is positive is not output. EC2: An operation mode in which the reference block at the bottom of the screen is handled. A motion vector whose Y component is positive is not output. EC3: An operation mode in which the reference block at the left end of the screen is handled. A motion vector with a negative X component is not output.

Therefore, in EC0-EC3, nine 4-bit signals {EC0, EC1, E as shown in FIG.
C2, EC3} must be entered. On the other hand, it is generally practiced to use a plurality of motion detection circuits in order to secure a larger search range.

FIG. 20 shows a case where six motion detection circuits are used.
It is a figure for demonstrating the case where each circuit detects a motion vector in a mutually different range. In this way, each circuit detects a motion vector in a different range, so that it is possible to detect a motion vector in a range that is six times as large as in the case where a single circuit is used. It should be noted here that the origin of operation of each circuit does not coincide with the center of the reference block (origin of motion vector).
That is, it has an offset 83 indicated by a middle arrow.

FIG. 3A is a diagram showing a reference block used in one of a plurality of motion detection circuits.
(B) is a diagram showing a search range used in one of the motion detection circuits using a plurality of. In FIGS. 3A and 3B, the block size is 4 × 4 pixels and the search range is 8 × 8 pixels, as in FIGS. 11A and 11B. Here, FIGS. 3A and 3B are shown in FIG.
The difference from (B) is that the pixel value of the reference block 41 is P
i, j (i = 0,1,2,3; j = 4,5,6,7).

This means that, as shown in FIG. 4, the motion vector origin 48 corresponding to the center position of the reference block does not coincide with the center of the search range 49. Therefore, in the case shown in FIG. 4, the value of the original motion vector is
Instead of (-1, -2), it becomes (+ 1, -4). Further, the motion vector limiting function at the screen edge described above should also be applied to the values (+1, -4) of this motion vector.

[0019]

However, conventionally, the expression of the motion vector has been performed by considering the origin 45 in the circuit operation corresponding to the center position of the search range as the origin. That is, the motion vector 4 in the case of FIG.
7 is a motion vector (−1, −2) in the circuit operation, and the offset 46 of the origin shown in FIG. 4 was not considered inside the motion detection circuit. In other words, it was the role of the external circuit to consider them. Therefore, the external circuit for controlling the screen edge is also complicated. As described above, the conventional motion detection circuit requires an external control circuit and an input terminal for adding a control signal,
There is a problem that the overall circuit scale increases.

The present invention has been made in view of the above-mentioned problems of the prior art, and it is an object of the present invention to reduce the size of the entire motion detecting device (circuit) using a so-called block matching method.

[0021]

In order to solve the above-mentioned problems of the prior art and to achieve the above-mentioned object, the image processing apparatus of the present invention has an operation indicating the position of a candidate block from an operation reference position. Motion vector generating means for generating the motion vector, the offset data indicating the operation reference position from the position of the reference block, and the operation motion vector are added to obtain the candidate from the position of the reference block. Addition means for calculating a reference motion vector indicating the position of the block, difference detection means for detecting a difference between the image data of the reference block and the image data of the candidate block, the detected difference and the calculated candidate Effectiveness of judging the validity of the motion vector in motion generated from the motion vector generating means, based on the reference motion vector of the block And a cross section.

Further, in the image processing apparatus of the present invention, preferably, the validity judging means judges the validity of the motion vector in operation by using the screen size for displaying the moving image data.

Further, the image processing apparatus of the present invention preferably further comprises a calculating means for calculating the absolute coordinates of the reference block from the screen size, and the validity judging means, and the detected difference, Based on the calculated reference motion vector of the candidate block and the absolute coordinates of the calculated reference block, the validity of the motion vector in motion from the motion vector generation means is determined.

Further, in the image processing apparatus of the present invention, preferably, the adding means adds the offset data, the motion vector in operation, and the absolute coordinates of the reference block to obtain the candidate block. An absolute coordinate is calculated, and the validity determining means determines the validity of the motion vector from the motion vector generating means based on the calculated absolute coordinates of the candidate block.

Further, the image processing method of the present invention generates an operation motion vector indicating the position of the candidate block from the operation reference position, and an offset indicating the operation reference position from the reference block position. The data and the motion vector on the operation are added to calculate a reference motion vector indicating the position of the candidate block from the position of the reference block, and the difference between the image of the reference block and the image of the candidate block is calculated. Then, the validity of the generated motion vector is determined based on the detected difference and the calculated reference motion vector of the candidate block.

Further, in the image processing method of the present invention, the validity of the motion vector in the operation is judged by using the screen size for displaying the moving image data.

[0027]

According to the image processing apparatus and the method thereof of the present invention, the motion vector generating means generates the motion vector indicating the position of the candidate block in the second image from the motion reference position. Then, in the adding means, the first
Offset data indicating the operation reference position from the position of the reference block in the image and the operation motion vector are added to obtain a reference motion vector indicating the position of the candidate block from the reference block position. It is calculated. Further, the difference detection means detects the difference between the image data of the reference block and the image data of the candidate block. Then, the validity determining means determines the validity of the motion vector in motion generated by the motion vector generating means based on the detected difference and the calculated reference motion vector of the candidate block. It

[0028]

DESCRIPTION OF THE PREFERRED EMBODIMENTS A motion detection circuit according to an embodiment of the present invention will be described below. First Embodiment In this embodiment, as described above with reference to FIG. 20, a plurality of motion detection circuits are used, and each circuit detects a motion vector in a different range. At this time, the motion detection circuit according to the present embodiment internally determines the validity of the motion vector. First, the processing contents of the motion detection circuit according to this embodiment will be described. Figure 3 (A)
Is a diagram showing a pixel pattern forming a reference block in the present embodiment, and FIG. 3B is a diagram showing a pixel pattern forming a search range in the present embodiment. In the present embodiment, the concept of terms such as reference block and search range is the same as that described in the section of the related art.

FIG. 4 is a diagram for explaining a method of calculating an original motion vector in the motion detection circuit according to this embodiment. As shown in FIG. 4, in the motion detection circuit according to the present embodiment, an offset 46 of the origin is given from an external device, for example, and the motion vector 44 in the circuit operation is added to the offset 46 of the origin to obtain the original motion vector 47 is calculated. Here, the original motion vector 47 is
As shown in FIGS. 4 and 5, it is a vector with the origin 48 of the motion vector on the previous screen 51 as a reference. The motion vector 44 for circuit operation is a vector with the origin 45 for circuit operation as a reference position for operation as a reference.
In the example shown in FIG. 4, the motion detection circuit according to the present embodiment is
The motion vector 44 in the circuit operation of (-1, -2) and (+
The offset 46 of the origin of 2, -2) is added, and (+
The original motion vector 47 of 1, -4) is calculated.

Next, as shown in FIG. 5, coordinates 39 of the reference block 40 in units of the number of pixels are given from an external device,
As shown in the following equation (3), the coordinate 3 of this reference block
9 and the original motion vector 47 calculated as described above.
By adding and, the coordinates (X
c, Yc) is calculated.

[0031]

(Equation 3)

At this time, the coordinates (Xc,
When the relations of the following equations (4) and (5) are established for Yc), the candidate block 43 is judged to be in the valid screen. That is, only at that time, the original motion vector 47 is effective.

[0033]

[Equation 4]

[0034]

(Equation 5)

In the above equations (4) and (5), M and N are values representing the horizontal and vertical sizes of the screen in pixel units, and similarly m and n are the horizontal and vertical sizes of one block, respectively. Is a value expressed in the number of pixels. In general, the values of m and n have already been determined by the standard of the motion detection circuit. For example, in the example shown in FIG. 4, m = n = 4
Is.

The specific configuration of the motion detection circuit according to this embodiment will be described below. FIG. 1 is a block diagram of a motion detection circuit according to this embodiment. As shown in FIG. 1, the motion detection circuit according to the present embodiment has input terminals 1, 2, 4,
5, 17, output terminals 19 and 20, a motion vector generator 3, a buffer 7, an adder 8, a residual calculation circuit 10, a comparator 13, and registers 15 and 16.

From the input terminal 1, the image data S in the search range
1 is input. The image data S2 of the reference block is input from the input terminal 2. The offset 46 of the origin is input from the input terminal 4. The coordinates 39 of the reference block are input from the input terminal 5. The screen size S17 is input from the input terminal 17.

The motion vector generator 3 generates the motion vector 44 for the circuit operation shown in FIGS. 4 and 5, and outputs the motion vector 44 for the circuit operation to the buffer 7, the adder 8 and the register 15. At this time, the motion vector 4 in the circuit operation generated by the motion vector generator 3 is generated.
4, each of the X and Y components can take five values from −2 to +2, and has 5 × 5 = 25 patterns. The motion vector generator 3 sequentially outputs 25 patterns of motion vectors 44 in circuit operation.

The buffer 7 temporarily stores the image data S1 in the search range input from the input terminal 1, and the value of the motion vector 44 in the circuit operation from the motion vector generator 3 in the image data S1 in the search range. The image data 9 of the candidate block selected according to is output to the residual calculation circuit 10.

The residual calculation circuit 10 inputs the image data 9 of the candidate block and the image data S2 of the reference block, calculates the residual 12 of these, and compares the residual 12 with the comparator 13.
And output to the register 16. The above formulas (1), (2)
As described above by using, the value of the motion vector 44 in the circuit operation when the value of the residual 12 becomes the minimum is the motion vector finally obtained.

The motion vector 44 in the circuit operation from the motion vector generator 3 is temporarily stored in the register 15,
The residual 12 from the residual calculation circuit 10 is temporarily stored in the register 16.

The adder 8 includes an offset 46 of the origin input via the input terminal 4, coordinates 39 of the reference block input via the input terminal 5, and the motion vector generator 3.
The motion vector 44 in the circuit operation is input, and the coordinate (Xc, Y
c) is output to the comparator 13. As described above, the coordinates (Xc, Yc) of the candidate block 43 are expressed by the above formula (4),
The motion vector is valid only when (5) is satisfied.

The comparator 13 uses the equations (4) and (5) described above.
It is determined whether the original motion vector 47 is valid based on the above, and the previous residual and the current residual output from the residual calculation circuit 10 are compared.

FIG. 2 is a flow chart for explaining the processing in the comparator 13. As shown in FIG. 2, in the comparator 13, the candidate blocks 4
Regarding “Xc” at the coordinate of 3, “0 ≦ Xc ≦ M−
It is determined whether or not the relation of "m" is established, and if it is established, the process of step S2 is executed, and if it is not established, the contents of the registers 15 and 16 are held as they are.

In step S2, the comparator 13 determines whether or not the relationship "0≤Yc≤N-n" holds for "Yc" at the coordinates of the candidate block 43, and if so, the process of step S3. Is executed, and if not satisfied, the contents of the registers 15 and 16 are held.

Further, in the comparator 13, in step S3, the relation "D <Dr" is satisfied for D indicating the residual 12 from the residual calculating circuit 10 and the data Dr indicating the residual held in the register 16. It is judged whether or not the condition is satisfied, and if so, the update command signal 1 is sent to the registers 15 and 16.
4 is output and the contents of the registers 15 and 16 are updated. On the other hand, in step S3, the comparator 13 sets “D <D
If the relation "r" is not established, the contents of the registers 15 and 16 are retained as they are.

By the processing in the comparator 13 described above, when invalid data outside the screen is input, the motion vector is detected only within the range of valid data.

When the update command signal 14 is input from the comparator 13, the registers 15 and 16 have their contents newly input from the motion vector generator 3 and the residual calculation circuit 10, respectively, and the motion vector 44 and the operation vector 44 in the circuit operation. Residual 12
Rewrite to the contents of.

The motion vector 44 for the circuit operation finally held in the register 15 is output as the motion vector S19 via the output terminal 19, and this motion vector S1
9 is used to compress the moving image data.

As described above, according to the motion detection circuit of the present embodiment, the motion vector is calculated in consideration of the origin offset 46 shown in FIG.
Therefore, the external device does not need to calculate the motion vector in consideration of the origin offset 46, and the external device is simplified and the number of input / output terminals of control signals in the external device and the motion detection circuit is reduced.
The overall size of the device can be reduced.

Second Embodiment FIG. 6 is a block diagram of a motion detection circuit according to this embodiment. 6, constituent elements given the same reference numerals as those in FIG. 1 are the same as the constituent elements described in the first embodiment. As shown in FIG. 6, in the motion detection circuit according to the present embodiment, the adder 21 is added, and the motion vector 44 in the circuit operation outputs the motion vector 44 in the circuit operation to the adder 21 and the input terminal 4 outputs the motion vector 44. The origin offset of 46 is output to the adder 21, and the adder 8 and the register 15 input the original motion vector 47 from the adder 21, except that the motion detection according to the first embodiment shown in FIG. It is the same as the circuit.

The adder 21 adds the origin offset 46 from the input terminal 4 and the motion vector 44 in the circuit operation from the motion vector generator 3 to obtain the original motion vector 47 as a result of the addition. It is output to the adder 8 and the register 15. That is, in the motion detection circuit of this embodiment, the original motion vector 47 is accumulated in the register 15, and the finally obtained original motion vector 47 is output via the output terminal 19.

Therefore, according to the motion detection circuit of the present embodiment, the original motion vector 47 is output from the output terminal 19, so that in the moving image compression processing device, the motion vector 44 and the offset of the origin in the circuit operation are offset. It is not necessary to calculate the original motion vector 47 from 46, and the moving image compression process can be performed using the original motion vector 47 from the motion detection circuit as it is. Therefore, in the motion detection circuit according to the present embodiment, the processing performed in the moving image compression processing device can be simplified, and the labor and time involved in the design of the moving image compression processing device can be reduced. by this,
The entire device (circuit) including the motion detection circuit and the moving image compression processing device can be downsized. Such an effect is particularly remarkable when the motion detection range is expanded by using a plurality of motion detection circuits as in the present embodiment.

Third Embodiment FIG. 7 is a block diagram of a motion detecting circuit according to this embodiment. In FIG. 7, the components denoted by the same reference numerals as those in FIGS. 1 and 6 are the same as the components described in the first and second embodiments. As shown in FIG. 7, the motion detecting circuit according to the present embodiment has the same structure as the motion detecting circuit according to the second embodiment shown in FIG.
3, the counter 23 is configured to calculate the coordinates 39 of the reference block from the screen size S17 from the input terminal 17.

That is, in the motion detecting circuit according to this embodiment, the coordinates 39 of the reference block are calculated by the built-in counter 23 using the screen size S17 without inputting from the external device. Therefore, according to the motion detection circuit of this embodiment, the number of input terminals can be reduced, and the whole device can be simplified and downsized.

The present invention is not limited to the above embodiments. For example, in the above-described embodiment, the case where a plurality of motion detection circuits are used as described above with reference to FIG. 20 and each circuit detects a motion vector in a range different from each other has been exemplified. The same can be applied to the case of detecting a motion vector by using the motion detection circuit of.

[0057]

As described above, according to the image processing apparatus and the method of the present invention, the so-called block matching method is used to detect the motion vector used in the compression processing of the moving image and the like. Without using the device, the motion vector can be calculated inside the device in consideration of the offset data of the operation reference point. for that reason,
In the external device, it is not necessary to calculate the motion vector in consideration of the offset data of the operation reference point, and the number of input / output terminals can be reduced and the entire device can be downsized and simplified. As a result, the labor and time required for designing the external device can be reduced.

[Brief description of drawings]

FIG. 1 is a configuration diagram of a motion detection circuit according to a first embodiment of the present invention.

FIG. 2 is a flowchart for explaining a process in the comparator shown in FIG.

FIG. 3A is a diagram showing a reference block used in one of a plurality of motion detection circuits, and FIG. 3B is a search range used in one of a plurality of motion detection circuits. FIG.

FIG. 4 is a diagram for explaining a motion vector when a plurality of motion vector detection circuits are used.

FIG. 5 is a diagram for explaining an example of each coordinate value in motion vector detection.

FIG. 6 is a configuration diagram of a motion detection circuit according to a second embodiment of the present invention.

FIG. 7 is a configuration diagram of a motion detection circuit according to a third embodiment of the present invention.

FIG. 8 is a diagram for explaining a moving image.

FIG. 9 is a diagram for explaining a motion vector.

FIG. 10 is a diagram for explaining motion detection using block matching.

11A is a diagram showing a pixel pattern forming a reference block, and FIG. 11B is a diagram showing a pixel pattern forming a search range.

FIG. 12 is a diagram for explaining the case where the motion vector is (−2, −2).

FIG. 13 is a diagram for explaining the case where the motion vector is (−2, −1).

FIG. 14 is a diagram for explaining a case where a motion vector is (−2, +2).

FIG. 15 is a diagram for explaining a case where the motion vector is (+2, +2).

FIG. 16 is a diagram for explaining an example of a motion vector detection range.

FIG. 17 is a diagram for explaining an example of motion detection at an image edge.

FIG. 18 is a diagram for explaining an example of a motion detection range at an image edge.

FIG. 19 is a diagram for explaining the relationship between the position of a reference block and an edge control signal.

FIG. 20 is a diagram for explaining an example of motion detection using a plurality of circuits.

[Explanation of symbols]

 1, 2, 4, 5, 17, 19, 20 ... Input terminal 3 ... Motion vector generator 7 ... Buffer 8, 21 ... Adder 10 ... Residual operation circuit 12 ... Residual 13 ... Comparator 15, 16 ... Register 23 ... Counter 39 ... Reference block coordinates 40 ... Reference block 44 ... Motion vector on circuit operation 46 ... Origin offset 47 ... Motion vector 48 ... Motion vector origin

Claims (6)

[Claims] 1. Image data of a reference block in a first image in moving image data and image data of a candidate block in a predetermined search range in a second image displayed before the first image. An image processing apparatus that calculates a motion vector indicating a positional relationship between the candidate block and the reference block based on a difference between the motion vector and a motion vector indicating a position of the candidate block from a motion reference position. Motion vector generating means for generating a vector, the offset data indicating the operation reference position from the position of the reference block, and the operation motion vector are added, and the candidate block from the position of the reference block And a difference between the image data of the reference block and the image data of the candidate block is detected. Minuteness detecting means, Effectiveness determination for determining the effectiveness of the motion vector generated by the motion vector generating means based on the detected difference and the reference motion vector of the calculated candidate block An image processing apparatus having a means. 2. The image processing apparatus according to claim 1, wherein the validity determining means determines the validity of the motion vector in operation by using a screen size for displaying moving image data. 3. A calculation means for calculating absolute coordinates of the reference block from the screen size is further provided, and the validity determination means is configured to detect the detected difference and a reference motion vector of the calculated candidate block. 3. The image processing apparatus according to claim 1, wherein the validity of the motion vector in the motion from the motion vector generating means is determined based on the calculated absolute coordinates of the reference block. 4. The adding means calculates the absolute coordinates of the candidate block by adding the offset data, the motion vector in operation, and the absolute coordinates of the reference block, and the validity determining means. The image processing apparatus according to claim 1, wherein the image processing apparatus determines the validity of the motion vector in motion from the motion vector generation means based on the calculated absolute coordinates of the candidate block. 5. Image data of a reference block in a first image in moving image data and image data of a candidate block in a predetermined search range in a second image displayed before the first image. An image processing method for calculating a motion vector indicating a positional relationship between the candidate block and the reference block based on a difference between the motion vector and a motion vector indicating a position of the candidate block from a motion reference position. Generating a vector, adding offset data indicating the operation reference position from the position of the reference block and the operation motion vector to obtain a reference indicating the position of the candidate block from the position of the reference block A motion vector is calculated, a difference between the reference block image and the candidate block image is detected, and the detected difference and the calculated candidate block are calculated. An image processing method for determining the validity of the generated motion vector on the basis of a reference motion vector. 6. The image processing method according to claim 5, wherein the validity of the motion vector in the operation is judged by using the screen size for displaying the moving image data.