JP3143494B2 - Motion vector apparatus and method - Google Patents

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 apparatus and method for detecting a motion vector of an image between screens.

[0002]

2. Description of the Related Art As a motion vector detecting method, there is a method called a block matching type.

[0003] The block matching type motion vector detection method is that (8 × 8) is used between the previous frame and the current frame.
Or, take a pixel of (16 × 16) or the like to make one block,
Matching is performed for each block, and a motion vector is given to each block of the current frame.

Next, the basic concept of the motion vector detection method will be described in detail.

[0005] A block of m horizontal pixels and n vertical pixels whose pixel position at the upper left corner is (i, j) on the current frame is B
_ij (m, n), and the block at the same position on the previous frame is represented by P _ij (m, n). At this time, the evaluation function of the error between B and P is

[0006]

[Outside 1] When x and y are changed from ± 0 to s (s is a parameter of a search range), the x and y are changed so that the smallest value of f is obtained from i and j. Vector.

In general, a color moving image is formed by dividing one frame into luminance data (Y) and chrominance data (Cr, Cb, etc.). The luminance data (Y) included is used.
In addition, the method of searching for a new vector centering on the vector obtained in the immediately preceding block or the surrounding block, or the correlation between the vector generated from each block and the block position is calculated, and the motion of the entire screen is calculated. A method of estimating is known.

[0008]

However, an image having an actual meaning (an image having information as a so-called picture such as a person, a landscape, a test pattern, etc., rather than a set of RGB luminance data) has many colors. , And a set of points having substantially the same hue means a single target.

In matching by blocks using luminance data, as is generally performed, a calculation error value can be locally minimized, but it is difficult to detect the same object in an image. is there.

Therefore, when trying to obtain the motion of the entire screen from individual vectors, or when trying to use a vector generated in a previous block or a processed peripheral block for vector search in the current block, a large error occurs. Will be invited.

In view of the above, an object of the present invention is to provide a motion vector detecting device and a method thereof capable of detecting the motion of the current screen image with respect to the other screen image more accurately than in the past.

[0012]

A motion vector detecting device according to the present invention comprises: input means for inputting color image data composed of luminance component data and color component data; and a current screen input by the input means. Detecting means for detecting a motion vector of the color image data of the current screen by comparing the color image data of the other screen with the color image data of another screen, wherein the detecting means converts the luminance component data and the color component data into The motion vector is detected using an evaluation function as a parameter, and the luminance component data and the color component data of the color image data input by the input means with respect to the luminance component data and the color component data in the evaluation function. And weighting is adaptively performed according to Further, the motion vector detecting method according to the present invention includes an input step of inputting color image data composed of luminance component data and color component data; Detecting a motion vector of the color image data of the current screen by comparing the motion vector with data of the current screen, wherein the detecting step uses the evaluation function having luminance component data and color component data as parameters. Detecting a vector, and adaptively weighting luminance component data and color component data in the evaluation function according to the luminance component data and color component data of the input color image data. And

[0013]

[0014]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment according to the present invention will be described below in detail with reference to FIGS.

FIG. 1 is a block diagram of a moving picture compression encoding apparatus using a vector detection circuit according to the present embodiment, and FIG. 2 is a block diagram showing a configuration of the vector detection circuit.

In FIG. 1, reference numeral 1 denotes a vector detection circuit which will be described in detail later, 2 denotes a frame memory for storing a previous frame, 3 denotes a subtraction circuit, 4 denotes a buffer memory, 5 denotes a quantization circuit, and 6 denotes lossless encoding. Circuit 7 is the quantization circuit 3
And 8 is an adder circuit that performs the reverse operation of.

The compression operation of the moving picture compression and encoding apparatus configured as described above will be described in detail.

The image data (brightness L ^* , chromaticity a ^* , b ^{* of} 8 bi each) of the current frame is transmitted via signal lines 101 and 102.
(represented by t) divides one frame into a plurality of blocks (for example, composed of 8 × 8 pixels), and is input to the vector detection circuit 1 in block units.

On the other hand, a block address signal is input to the frame memory 2 via the signal line 103.

In accordance with the block address signal, image data in a vector search range around the block is read from the frame memory 2 and input to the vector detection circuit 1 via the signal line 104.

The vector detection circuit 1 outputs, to the signal lines 105 and 106, block data evaluated as being optimum from a predetermined search range.

Next, the current image block delayed by a predetermined time by the buffer memory 4 is input to the subtraction circuit 3 via the signal line 107, and at the same time, the value of the optimum block is transmitted to the subtraction circuit 3 via the signal line 105. Is entered.

The subtraction circuit 3 subtracts the values of the input current image block and the optimum block, and inputs the difference value data to the quantization circuit 5 for quantization.

The quantized data is input to a lossless encoding circuit 6 and an inverse quantization circuit 7.

The lossless encoding circuit 6 assigns a lossless code to the quantized data, and the lossless encoded data is transmitted to the signal line 10.
8 is output.

The inverse quantization circuit 7 reproduces the difference value, and the addition circuit 8 adds the difference value and the optimum block data input via the signal line 109. The added data is stored in the frame memory 2.

Next, the vector detection circuit 1 shown in FIG.
Will be described in detail with reference to FIG.

First, the previous frame image data of the search area is input from the frame memory 2 (see FIG. 1) to the memory 20 via the signal line 104. The current image data is signal line 1
02 to the evaluation circuit 21.

The evaluation circuit 21 determines the parameters of the evaluation function in the block matching based on the data distribution of the original block, and transmits the result to the arithmetic circuit 22 via the signal line 201. The method for determining the parameters of the evaluation function will be described later.

The arithmetic circuit 22 outputs a vector to the target block to be matched to MPX via a signal line 202.
(Multiplexor) 23.

The MPX 23 generates an address signal in accordance with the input vector. The address signal is input to the memory 20 via a signal line 203, and is read from the memory 20 by the address data via a signal line 204. The output block data is received, and the block data is input to the arithmetic circuit 22 via 205.

The arithmetic circuit 22 having received the two block data calculates the error between the blocks using the following evaluation function equation.

[0033]

[Outside 2] (L, a, and b are parameters determined by the evaluation circuit 21 and will be described later. Subscripts have the same meaning as in the conventional example.
An error value between blocks and vector data of a matched block are input from a calculation circuit 22 to a comparison circuit 24 via a signal line 206 from a calculation circuit 22.

The comparison circuit 24 selects the smallest one and the vector at that time from among a large number of block matching error data calculated for one current block. When the block matching for one current block is completed, the arithmetic circuit 22 outputs the vector output circuit 25
Via the signal line 208. In response to this, the vector output circuit 25 reads the vector data stored in the comparison circuit 24, outputs the vector data to the outside via the signal line 106, and reads the block data to the MPX 23 via the signal line 209. Input the signal. The MPX 23 follows the memory 2
Read the best matching block data from 0,
The block data is output to the outside via a signal line 105.

Next, an evaluation method in the evaluation circuit 21 shown in FIG. 2 will be described.

The input block is represented by 8 bits each of L ^* , a ^* , and b ^* . In L ^* , L ^* = 255.
Is white, and in a ^* and b ^* , a ^* = 127 (same for b ^* ) is colorless in the axial direction, and a ^* = 0, a ^* = 255 (same for b ^* ) is the most colorful in both directions of the axis. The degree is high. The block is composed of (m × n) pixels, and the block position in the frame is represented by (i, j) and the pixel position in the block is represented by (p, q). L ^* , a ^* , b to be input
^* Each block will be referred to as L _ij , A _ij , B _ij, and in particular, for describing the pixel position therein, for example, L _ij
ij (p, q).

The evaluation circuit 21 calculates A according to the following equation.
For each pixel in each of the (a ^* block) and B (b ^* block), a variance value is calculated based on the data value 127, and the values are defined as VA and VB.

[0038]

[Outside 3]

[0039]

[Outside 4] According to the calculation result by the above equation, the larger one of VA and VB is defined as VC. Using these VA, VB, and VC, the coefficients, a, b, and l in the above-described evaluation function formula are determined. The relationship between VA, VB, VC and a, b, l follows the formula shown below.

A = fab (VA) b = fab (VB) l = fl (VC) where fl and fab are predetermined functions, and the functions fl and fab
FIG. 3 shows an example of the fab shape.

The values of the parameters a, b, and l determined here are output from the signal line 201 in FIG.

In this embodiment, lightness data (L ^* ) and chromaticity data (a ^* , b ^* ) are used.
It is obvious that the same can be realized by using Cb, YIQ, LUV or the like.

[0043]

As described above, according to the present invention,
A motion vector is detected using an evaluation function having luminance component data and color component data as parameters. Luminance component data of color image data input with respect to the luminance component data and the color component data in the evaluation function. Since weighting is adaptively performed in accordance with the image data and the color component data, highly reliable motion vectors can be detected for various images.

[Brief description of the drawings]

FIG. 1 is a block diagram of a motion vector detection device according to an embodiment.

FIG. 2 is a block diagram of a vector detection circuit used in the device of FIG.

FIG. 3 is a diagram illustrating shapes of functions fl and fab.

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 vector detection circuit 2 frame memory 3 subtraction circuit 4 buffer memory 5 quantization circuit 6 lossless encoding circuit 7 inverse quantization circuit 8 addition circuit 21 evaluation circuit 22 operation circuit 24 comparison circuit 25 vector output circuit

──────────────────────────────────────────────────続 き Continued on the front page (58) Field surveyed (Int.Cl. ⁷ , DB name) H04N ^7/ 24-7/68 H04N 11/00-11/22

Claims (3)

(57) [Claims] 1. An input unit for inputting color image data composed of luminance component data and color component data, and comparing the color image data of the current screen input by the input unit with the color image data of another screen. Detecting means for detecting a motion vector of the color image data of the current screen, wherein the detecting means detects the motion vector using an evaluation function having luminance component data and color component data as parameters. Adaptively weighting the luminance component data and the color component data in the evaluation function according to the luminance component data and the color component data of the color image data input by the input means. Motion vector detection device. 2. The motion vector detecting apparatus according to claim 1, further comprising coding means for performing motion compensation predictive coding of the color image data of the current screen using the motion vector detected by said detecting means. 3. An inputting step of inputting color image data composed of luminance component data and color component data, and comparing the input color image data of the current screen with color image data of another screen. A detecting step of detecting a motion vector of the color image data of the current screen, wherein the detecting step detects the motion vector using an evaluation function having luminance component data and color component data as parameters, and A motion vector detection method characterized by adaptively weighting luminance component data and color component data in a function according to the luminance component data and color component data of the input color image data.