KR0180162B1 - Apparatus and method for processing a motion vector - Google Patents

Abstract

The present invention relates to a motion vector processing apparatus and a method thereof. In the motion estimation method of performing a high resolution motion estimation after performing a low resolution motion estimation, the apparatus of the present invention includes a motion vector generated at the time of high resolution motion estimation for the entire search region. A control unit 100 for outputting a control signal so as not to deviate; The motion estimator 200 is configured to limit and process a motion vector by performing an edge control function according to a control signal output from the controller 100. The method of the present invention provides a low resolution motion estimation method in a decimated image. The edge control function of the motion estimation dedicated chip is used to ensure that the motion vector generated during the high resolution motion estimation that searches only a part of the original image does not leave the entire search area. By limiting this, an error can be prevented from occurring, which has the advantage.

Description

An apparatus and the method for processing a motion vector

1 is a block diagram according to a two-step motion estimation method.

2 is an exemplary diagram showing motion estimation in low resolution and high resolution images.

FIG. 3A is a block diagram schematically illustrating a two-step motion estimation scheme. FIG.

(b) is an illustration showing an original image and a decimated image.

(c) is an exemplary diagram illustrating high resolution motion estimation using a motion vector obtained at the time of low resolution motion estimation.

4 is a block diagram of a motion vector processing apparatus according to the present invention.

5 is a block diagram of an embodiment of a motion estimation unit according to the present invention.

FIG. 6A illustrates a search window of a 4n x 8 reference block by a dedicated chip for motion estimation.

(b) shows a 4n x 8 reference block.

7A to 7F are timing charts showing timings of an initialization sequencer using a motion estimation dedicated chip.

8A to 8F are timing diagrams showing timing of a block sequencer using a motion estimation dedicated chip.

9A to 9G are timing charts showing output timings by a motion estimation dedicated chip.

FIG. 10 is an exemplary diagram illustrating one frame for explaining an edge control function of a motion estimation dedicated chip.

* Explanation of symbols for main parts of the drawings

100: control unit 200: motion estimation unit

210: data former 220: 256 processor array

230: error RAM 240: error comparator

250: I / O bus 260: controller

The present invention relates to a motion vector processing apparatus and a method thereof, and more particularly, to a motion vector processing apparatus and method for limiting a motion vector from leaving the entire search area by using an edge control function of a motion estimation dedicated chip in an image encoder. will be.

Video encoders that use the Moving Picture Experts Group-2 (MPEG-2) method, such as HDTV (High Definition TV) or video phones, undergo various data compression processes to reduce the data to be transmitted to the video decoder. One method of the data compression process is motion estimation.

In general, motion estimation is an estimation of a motion vector indicating how much the pixels of the current frame move with respect to the pixels of the previous frame in a continuous image signal. Instead, it refers to an image data compression technique that reduces the amount of image data transmitted by transmitting the motion vectors as described above.

The motion estimation method in each method is as follows.

1) ATRC method (ADTV: Advanced TV)

The motion estimation method used for ADTV dp adopts a block matching algorithm (BMA) with half-pixel accuracy, which is an algorithm proposed in MPEG.

The block matching algorithm (BMA) assumes that the motion of the screen is moved horizontally or vertically in parallel, so that the block image of the frame in which the movement occurs (i.e., the current frame) is positioned at a position in the frame (i.e., the previous frame) before the movement occurs. It is a method of estimating the motion vector through the position by estimating whether the block image is the best match.

Motion estimation in ADTV refers to the process of calculating a two-dimensional vector representing the degree of motion. These vectors are called motion vectors.

Two types of motion estimation are used in ADTV, one for forward motion estimation using the past screen to make the current screen, and the other for backward motion estimation using the future screen to make the current screen. To do.

Both methods find the motion vector by finding the block with the least distortion in the predetermined search interval with M * N.

Common methods of measuring error (or distortion) include Mean Square Error (MSE) and Mean Absolute Error (MAE). ) Is mainly used.

Half-pixel motion estimation consists of two steps. The first search is to obtain an integer-pixel motion vector, and the half-pixel motion vector is centered around the pixel-by-pixel motion vector first obtained by the second search. Will be obtained.

In ADTV, the input video is divided into several types and divided into frames. There are three types of I-frame, P-frame, and B-frame, and motion estimation for each frame. Looking at it as follows.

i-frames do not have motion estimation, so there is no motion vector, but they are used for motion estimation of other frames. P-frames use the previous I-frame or P-frame to perform forward motion estimation and another P-frame or P-frame. Forward motion estimation and backward motion estimation are then performed, and the forward motion corrected frame and the backward motion compensated frame are added together to create an interpolated frame divided into two. The displaced block differential is chosen to determine the motion vector.

When the smallest value of the DBD is interpolated, both the forward motion vector and the backward motion vector are transmitted.

2) G.I company digicipher

The motion estimation method in G.I's Digicipher system proposal is similar to the block matching algorithm (BMA) used in ATRC's ADTV described above, but is less than ADTV because it does not estimate half-pixel motion in accuracy.

In other respects, there is no B-frame of ADTV by performing forward motion estimation without backward motion estimation. That is, the Digicipher frame consists of only I-frames and P-frames.

The unit of motion estimation is a macro block. The macro block is composed of a 4 * 2 Y (luminance signal) block, one U (color difference signal) block, and one V (color difference signal) block. Have.

3) MIT method (ATVA-Progressive HDTV)

The motion estimation method proposed by MIT is based on the spatial-temporal constraint, which assumes that image displacement occurs at a constant rate between frames and correlates the displacement rate with the motion vector. It is a method to estimate the motion vector by finding the spatial-temporal constraint, which is a differential equation representing the relationship, and finding its root.

This method assumes a uniform velocity transmission and is not valid for multi-object movements with object rotation, camera zoom, moving object transmission regions, or other motion vectors.

Even if the motion is assumed to be uniform within a small interval and motion estimation is suppressed in an area where the motion estimation is uncertain, the temporal correlation is greatly reduced by motion estimation and compensation based on spatial-temporal constraints. do.

Compared with the block matching algorithm (BMA), this method greatly reduces the amount of computation depending on the block size and performs well with or without noise.

The motion vector is estimated from the luminance signal and the same motion vector is used for the luminance signal Y and the chrominance signal U and Y, one motion vector is obtained for every 32 * 32 pixel block.

Prediction is performed on the current frame from the previous coded frame and the motion vector, and a difference between the current frame and the predicted code is calculated. This difference is referred to as a motion compensated residual.

This is obtained for each of the three parts Y, U and V and when the image has relatively large enough energy (when there is a picture change) relative to the picture frame, motion compensation is invalidated and the picture frame itself is encoded.

4) Zenith and ATT (DSC-HDTV)

Due to the nature of the image, the pixel values of adjacent pixels are generally composed of very close values because the pixels in the lines representing the boundary and contour of the object are used to display the same object.

Zenith and ATT used a hierarchical motion estimation method that effectively estimates motion by eliminating unnecessary redundancy of frame image data by using the characteristics of these images.

FIG. 1 is a block diagram according to a two-step motion estimation method, and FIG. 2 is a diagram showing motion estimation in low resolution and high resolution images. In order to increase transmission channel efficiency, low resolution motion estimation is performed by Zenith and ATT. Two-stage motion estimation of motion estimation and fine motion estimation is proposed. The proposed method is described in detail as follows.

First, a filter is used to remove redundancy between the frame video signal and the previous video signal. As a result, the resolution is reduced by 1/2 each in the horizontal and vertical directions.

Coarse motion estimation from low resolution video signals with an overall reduction of 1/4 is an estimation error that is the absolute value of the pixel value difference between the current frame and the previous frame in units of 32 * 16 (horizontal * vertical) pixels. In this case, the smallest of the estimation errors obtained by using only the luminance component and moving the previous frame block by the search interval becomes the motion vector of the block.

The block matching in the hierarchical low resolution image as described above reduces the search interval and the block size by a quarter compared to that used in the original high resolution image, thereby significantly reducing the calculation amount.

High resolution motion estimation uses motion vectors obtained from low resolution images to replace them with motion vectors in the original high resolution images and subdivides them into high resolution motion vectors.

That is, the 16 * 8 block size and the low resolution motion vector in the low resolution image are each reduced by 1/2 in the horizontal and vertical directions of the original resolution image, so double them all to obtain the block size and the motion vector of the original image.

This is divided into 8 * 8 small block sizes, and each small block is moved back from the low resolution motion vector by the search interval to obtain a high resolution motion vector as shown in FIG.

The motion vector encoder estimates the screen of the next frame using motion vectors estimated at low resolution and high resolution, respectively, and selects a motion vector corresponding thereto.

This is for limiting the bit rate in compressing the motion vector, and depending on the margin of the channel band width, it may send only a low resolution motion vector or a low resolution and high resolution motion vector.

There are many cases where a motion estimation chip is used when making an MPEG-2 based video encoder, since the possible search range of the motion estimation dedicated needle is often smaller than the system's required search range. It must be processed in parallel with a large amount of chips for motion estimation.

In other words, in the motion estimation method used in the MPEG-2 method, the larger the search area for the motion estimation becomes, the larger the calculation amount becomes. Since there is no dedicated motion estimation chip that satisfies the search area, a large amount of the motion estimation dedicated chip is used. System must be designed to satisfy the full search range.

However, the use of a large amount of dedicated motion estimation chips is a big problem in the implementation of the system, and thus the number of chips is reduced by various methods.

FIG. 3 (a) is a block diagram schematically illustrating a two-step motion estimation method. In the first step, a low-resolution motion estimation method for obtaining a rough motion vector in a decimated image from an entire image is shown. In the second step, a high-resolution motion estimation (fine ME) is performed, which returns only to the original image and searches only a partial region based on the motion vector obtained in the first step. The motion estimation method is adopted.

Looking at the two-step motion estimation method in more detail as follows.

(B) of FIG. 3 is an exemplary view showing the original image and the decimated image. In order to reduce the number of motion estimation dedicated chips (eg, STI3220 of SGS-Thomson), that is, to reduce the amount of motion estimation calculations. The original image is decimated to reduce the size of the original image. The motion estimation is performed with the horizontal direction reduced to 1/4 and the vertical direction reduced to 1/4.

FIG. 3 (c) shows an example of high resolution motion estimation using a motion vector obtained at low resolution motion estimation, in which a denotes an image of an anchor frame and an image of a current frame in a decimated image. The motion vector is obtained by performing the motion estimation, and b is the motion vector obtained from the decimated image, which is enlarged and then applied to the original image to perform motion estimation again.

In general, a search area used in an HDTV or the like has several pieces -128 to +127 -64 to +63 in integer pixel units.

As shown in (a) of FIG. 3, when the low resolution motion estimation is performed on the decimated image in the horizontal and vertical directions each quarter, the search area is horizontal -32 to +31 and vertical -16 to You get +15.

When performing high resolution motion estimation using the motion vector obtained from the reduced search area as described above, the motion vector is enlarged 4 times horizontally and vertically, respectively, and then a certain section, ie horizontal -8, is centered on the vector in the original normal screen. The motion estimation is performed again for the search areas of ~ +7 and vertical -8 ~ +7.

In general, no problem occurs, but the following occurs.

Considering only the horizontal direction, the motion vector obtained when estimating low resolution motion is -32. Alternatively, in case of +31, the range of -8 to +7 is searched again based on -128, -124, and +124, which are values multiplied by 4 when the high resolution motion is estimated.

In this case, since the result of the high resolution motion estimation occurs from -129 to -136 and +128 to +132, an error occurs out of the range of -128 to +127, which is the maximum search range defined by the system. There is a problem. This problem occurs equally for vertical motion vectors.

Accordingly, the present invention has been made to solve the above problems, and provides a motion vector processing apparatus and method using the edge control function of the motion estimation dedicated chip to limit the motion vector from leaving the entire search area. There is a purpose.

In the motion vector processing apparatus of the present invention for achieving the above object, in the motion estimation method of performing high resolution motion estimation after low resolution motion estimation, the motion vector generated at the time of high resolution motion estimation does not leave the entire search region. A control unit for outputting a control signal to the control unit; And a motion estimator configured to limit and process a motion vector by performing an edge control function according to the control signal output from the controller.

In addition, the motion vector processing method of the present invention for achieving the above object, in the motion estimation method of performing high resolution motion estimation after low resolution motion estimation, the motion generated after performing the low resolution motion estimation in the decimated image Based on the vector, the motion vector generated during the high resolution motion estimation that searches only a part of the original image is limited using the edge control function of the motion estimation dedicated chip so that the motion vector does not leave the entire search area. do.

Hereinafter, with reference to the accompanying drawings will be described in detail with respect to the present invention.

4 is a block diagram of a motion vector processing apparatus according to the present invention. In the motion estimation method in which high resolution motion estimation is performed after low resolution motion estimation is performed, the apparatus of the present invention performs a full search for a motion vector generated at the time of high resolution motion estimation. A control unit 100 for outputting a control signal so as not to leave the area; The motion estimator 200 is configured to limit and process a motion vector by performing an edge control function according to a control signal output from the controller 100.

5 is a block diagram of an embodiment of a motion estimator according to the present invention, in which the motion estimator 200 receives an x-port and an 8-bit pixel unit for receiving a reference block in 8-bit pixel units. A data generator 210 comprising A, B, and C ports for receiving search window data; 256 processor arrays 220; An error RAM 230 for storing an error of 16 bits; An error comparator 240 for comparing the errors; As a result of comparing the errors in the error comparator 240, the input and output bus 250 for transmitting the address and the error for accessing the motion vector having the minimum error and the error RAM 230; And , , , , The controller 260 is configured to input signals to control various operations and to generate timing of MVSYNC, DSYNC0, and DSYNC1 signals.

In addition, the motion vector processing method according to the present invention is a motion estimation method in which high resolution motion estimation is performed after low resolution motion estimation, wherein the original image is focused on a motion vector generated after low resolution motion estimation is performed on the decimated image. In order to prevent the motion vector generated during the high resolution motion estimation that searches only a part of the region from leaving the entire search area, the edge control function of the motion estimation dedicated chip is limited.

Subsequently, with reference to the accompanying drawings, the operation of the motion vector processing apparatus and the motion vector processing method of the motion vector processing apparatus according to the present invention will be described in detail by using the STI3220 manufactured by SGS-THOMSON, which is one of the motion estimation dedicated chips. Shall be.

As shown in FIG. 4, in the case of a motion estimation method in which high resolution motion estimation is performed after low resolution motion estimation, the control unit 100 prevents the motion vector generated at the time of high resolution motion estimation from leaving the entire search region. Outputs a control signal.

The motion estimator 200 restricts and processes a motion vector by performing an edge control function according to the control signal output from the controller 100.

Referring to Figure 5, Is a signal for resetting the motion estimation dedicated chip, Is the output enable signal, Is a timing signal requiring the start of a sequence, EC [3: 0] is a 4-bit data signal for edge control, Is low, the block's vertical size is 8 pixels, high means the block's vertical size is 16 pixels, Is a signal indicating that the block size is low (8 * 8 or 16 * 16) and low is rectangular.

FIG. 6 (a) shows a search window of a 4n x 8 reference block by a dedicated chip for motion estimation, and FIG. 6 (b) shows a 4n x 8 reference block. The 4n x 8 reference block of the motion estimation dedicated chip is composed of 4n pixels in the horizontal direction and 8 pixels in the vertical direction as shown in (b) of FIG. 6, and the search window of the reference block is shown in FIG. As shown in (a), the SWA1, SWB1, SWC1, SWA2, SWB2, and SWC @ sub-windows formed of 8 rows and 15 columns are divided into A ports in the initialization sequence. SWA1 is used as the port, SWB1 is used as the B port, and SWC1 is entered as the C port. SWA2 is used as the A port, SWB2 is used as the B port, SWC2 is used as the C port, and reference block is used as the X port. Are input respectively.

FIG. 7 is a timing diagram showing the timing of an initialization sequence by a motion estimation dedicated chip. The SI3220, a motion estimation dedicated chip, has an operation mode consisting of an initialization sequence and a block sequence. As shown in (a) to (e), 15 rows of SWA1, SWB1, SWC1 blocks are loaded during 128 clock cycles.

At this time, the beginning of the sequence The signal starts low for about one clock, not loaded during the first eight clocks, and the chip is initialized internally, followed by 120 pixels (8 x 15 = 120) of data per port for 120 clocks.

The order in which data is loaded is from the first column to the 15th column sequentially while loading in the vertical direction.

At this time, SWA1 and SWB1 are 8 pixels in the vertical direction, but SWC1 and SWC2 are 7 pixels. Since 8 pixels are input for timing, a column is additionally loaded as shown by a dotted line in FIG. It is excluded from the motion vector calculation.

(A) to (e) of FIG. 8 are timing diagrams showing the timing of the block sequence by the motion estimation dedicated chip. Loading, If the signal is low and the reference block is square (eg 8x8, 16x16), then at the beginning of the sequence It is not essential that the signal goes low, but if the reference block is not square ( = High), at the beginning of the sequence It is essential that the signal goes low.

(A) to (g) of FIG. 9 are timing diagrams showing output timings by a motion estimation dedicated chip. As shown in FIG. After the motion vector and the error of the candidate block having? Are obtained, they are output through the input / output bus IOB at the timing shown in Fig. 9 (a) and bet (g).

That is, in the M + 30th clock cycle of the next sequence after the block sequence is performed, as shown in (c) of FIG. 9, the output enable signal ( Is high, as shown in (d) of FIG. When the signal goes low, an 8-bit motion vector (MV) is loaded on the I / O bus. The highest 8 bits (MSB) out of the 16-bit error are output at the timing when the signal goes low. At the timing of becoming low, the lower 8 bits LSB of the error are output.

As such, the motion vector and the error value of the previous block sequence can be riddled through the I / O bus while the block sequence is being executed, and after outputting a specific address to the I / O bus, the error RAM of the corresponding address is read to read the error data. Can come.

FIG. 10 illustrates an example of a frame for explaining an edge control function of a motion estimation dedicated chip. Referring to FIG. 10, an edge control function of ST1 3220 (SGS-Thomson), which is a motion estimation dedicated chip, is described in detail with reference to FIG. Let's take a look.

If the search window is partially outside the actual frame, the number of candidate blocks is less than 256, and the useful motion vector is also reduced.

The motion estimation dedicated chip STI3220 calculates motion vectors within the useful area in consideration of the above.

There are nine different cases depending on whether the search window is located at one of the edges or at one of the four corners of the frame.

The Motion Estimation Dedicated Chip (STI3220) has four pins (EC0, EC1, EC2, EC3) for edge control, which are all latched in the circuit during the first cycle of every block sequence. It is related to the search window loaded during this cycle.

Each pin is associated with one edge of the frame (left, right, top or bottom), and making one pin high during the first cycle of the block sequence means that the current search window (loaded into the circuit during the partial block sequence) Means that the search window is off the edge of the frame associated with the pin and the useful motion vectors for the current search window are reduced.

The most common case is when the search window is completely contained within the frame, where all four pins are low.

The EC0 pin is related to the top edge of the frame. When this pin is made high, the useful area of the motion vectors is reduced to the area of motion vectors with positive or zero vertical components, and the EC1 pin is related to the right edge of the frame. Making the pin high reduces the useful area of the motion vectors to the area of the motion vectors with a negative or zero horizontal component, and the EC2 pin is related to the bottom edge of the frame. The EC3 pin is related to the left edge of the frame, and when this pin is made high, the useful area of the motion vectors is a motion vector with a positive or zero horizontal component. Is reduced to the area.

If the search window is out of the corner of the frame, two edges should be selected.

Table 1 below summarizes all combinations of edges, with unrealistic combinations marked with *.

Looking at the edge with reference to Figure 10 has nine cases as follows.

① oral which cannot have motion vector in the negative direction horizontally and vertically

② cannot have a motion vector in the negative direction vertically

③ When it cannot have a motion vector in a positive direction horizontally and a negative direction vertically

④ cannot have a motion vector in the negative direction horizontally

⑤ When there is no restriction on the range of motion vectors

⑥ cannot have a motion vector in the positive direction horizontally

⑦ It cannot have motion vector in horizontally negative direction and vertically positive direction

⑧ cannot have a motion vector in the positive direction vertically

⑨ If it is impossible to have a motion vector in a positive direction horizontally or vertically, the edge control function of the motion estimation dedicated chip shown in Table 1 is used to limit the motion vector to a certain range as described above. In the motion estimation method in which high resolution motion estimation is performed after low resolution motion estimation (coarse ME) and the high resolution motion estimation is performed on the outer edge of a frame image, the motion vector generated during the high resolution motion estimation is a search region. It is applied to the part that needs to be limited so as not to escape.

That is, if the motion vector is -32, -31 or 31 in the horizontal direction or -16, -15 or 15 in the vertical direction (refer to Table 2 below) for the low resolution motion estimation, the motion is estimated when the high resolution motion is estimated. Dedicated chip edge control can be used to limit the motion vector to always leave the search range.

According to the motion vector values of Table 2, the edge control function of the motion estimation dedicated chip (SGi-Thomson, Inc. STi 3220) is specifically applied with reference to Table 10 and Table 1 as follows.

① If the motion vector (i, j) in low-resolution motion estimation is i = -32 or -31 and j = -16 or -15, it cannot have a horizontal and vertical motion vector in the high resolution motion estimation. In this case, if the EC0 pin and the EC3 pin of the motion estimation dedicated chip are made high, the motion vector is limited to the search area.

② If the motion vector (i, j) in the low resolution motion estimation is -30 i 30 and j -16 or -15, since the motion vector in the negative direction cannot be vertically obtained when estimating the high resolution motion, Making the needle's EC0 pin high will limit the motion vector into the search range.

(3) When the motion vector (i, j) in low resolution motion estimation is i 31 and j -16 or -15, the motion vector in the high resolution motion and the vertical direction in the high resolution motion cannot be negative. Therefore, if the EC0 pin and EC1 pin of the motion estimation dedicated chip are made high, the motion vector is limited to the search range.

④ When the motion vector (i, j) in low resolution motion estimation is i -32 or -31 and -14 j, it is because it is not possible to have the motion vector in the negative direction horizontally when estimating the high resolution motion. If we make the EC3 pin high, the motion vector is confined to the search area.

⑤ If the motion vector (i, j) in the low resolution motion estimation is -30 i 30 and -14 j 14, since there is no limitation in the range of the motion vector to be obtained for the high resolution motion estimation, the edge control of the dedicated chip for motion estimation The pins EC0, EC1, EC2, and EC3 do not need to be high.

⑥ When motion vector (i, j) in low resolution motion estimation is i 31 and -14 j 14, EC1 pin of motion estimation dedicated chip is because it cannot have horizontally positive motion vector in high resolution motion estimation. Makes high, the motion vector is confined to the search area.

⑦ When the motion vector (i, j) in low resolution motion estimation is i -32 or -31 and j 15, the negative direction is horizontally when estimating the high resolution motion and the negative direction is horizontal when estimating the motion in the vertical direction. In this case, since the motion vectors in the positive direction cannot be vertically placed, when the EC2 and EC3 pins of the motion estimation dedicated chip are made high, the motion vectors are limited to the search area.

⑧ Motion vector (i, j) in low resolution motion estimation rk -30 i 30 and j 15 means EC2 pin of motion estimation dedicated chip because it is not possible to have vertical motion vector vertically in high resolution motion estimation. Makes high, the motion vector is confined to the search area.

⑨ If motion vector (i, j) in low resolution motion estimation is i 31 and j 15, EC1 pin of the motion estimation needle is not available because it is not possible to have the motion vector in the horizontal and vertical direction in high resolution motion estimation. And bringing the EC2 pin high, the motion vector is confined to the search area.

As described above, in the present invention, when high resolution motion estimation is performed using the motion vector generated at the low resolution motion estimation, the motion vector is limited to the edge control function of the motion estimation unit at the time of low resolution motion estimation. Since the motion vector is not generated outside the entire search area, an error can be prevented from occurring.

Claims (4)

A motion estimation method for performing high resolution motion estimation after low resolution motion estimation, the motion estimation method comprising: a control unit (100) for outputting a control signal so that a motion vector generated at the time of high resolution motion estimation does not leave the entire search range; And a motion estimator (200) configured to limit and process a motion vector by performing an edge control function according to a control signal output from the controller (100). The method of claim 1, wherein the motion estimator 200 includes an X port for receiving a reference block in 8-bit pixel units and an A, B, and C port for receiving search window data in 8-bit pixel units, respectively. Data generator 210; 256 processor arrays 220; An error RAM 230 for storing an error of 16 bits; An error comparator 240 for comparing the errors; An input / output bus 250 for transmitting a motion vector having a minimum error, an address for accessing the error RAM 230, and an error as a result of comparing the errors in the error comparator 240; And , EC [3: 0], , Input signals to control various operations A motion vector processing device, comprising: a controller (260) for generating timing of a signal. A motion estimation method for performing high resolution motion estimation after low resolution motion estimation, wherein the high resolution motion estimation searches only a partial region of the original image centered on a motion vector generated after performing low resolution motion estimation on a decimated image. A motion vector processing method for restricting the motion vector generated at the time using the edge control function of the motion estimation dedicated chip so that the motion vector generated at the time does not leave the entire search range. 4. The method of claim 3, wherein when the motion vector in the low resolution motion estimation is i -32 or -31 and j -16 or -15, the motion vector in the high resolution motion estimation cannot be horizontally or vertically negative. When the motion vector in the low resolution motion estimation is -30 i 30 and j -16 or -15, the motion vector in the low resolution motion is limited to have no vertical motion vector in the low resolution motion estimation. If the motion vector is i 31 and j -16 or -15, it is restricted to have a motion vector in a horizontally positive direction and a vertically negative direction when estimating a high resolution motion, and the motion vector in the low resolution motion estimation is i − In the case of 32 or -31 and -14 j 14, it is restricted to have a motion vector in the negative direction horizontally when estimating high resolution motion. If the motion vector in the high-resolution motion estimation is i 31 and -14 j 14, the motion vector in the low-resolution motion estimation is limited to i -32 or-in high resolution motion estimation. If it is 31 and j 15, it is limited to have a high resolution motion vector, and if the motion vector in low resolution motion estimation is -30 i 30 and j 15, it can have a vertically positive motion vector in high resolution motion estimation. And when the motion vector in the low resolution motion estimation is i 31 and j 15, limiting the motion vector in the high resolution motion estimation to have no horizontal and vertical motion vectors.