JPH10136385A - Luminance change compensation method for moving image, moving image coder, moving image decoder, recording medium recording moving image coding or decoder program and recording medium recording coded moving image data - Google Patents

Abstract

PROBLEM TO BE SOLVED: To attain more efficient coding in the case of coding a moving image having a global luminance change. SOLUTION: A luminance change detection section 5 uses each luminance of 16×16 pixels in a block Bi of an N-th frame original image and each luminance of 16×16 pixels at a location deviated by a motion vector Vi in an (N-1)th frame original image to calculate two parameters DC(i), DB(i) in the block Bi. A frequency calculation section 6 obtains the frequency of the DC(i), DB(i) for one frame and decides sets of parameters with a highest frequency as parameters DC, DB denoting a global luminance change between the N-th frame original image and the (N-1)th frame original image. A luminance change compensation section 7 corrects the luminance (x) of each picture element of the (N-1)th frame reproduction image by using the parameters DC, DB as perameters DC.x+DB.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an encoding process for efficiently transmitting and storing digital moving images, and a global luminance change amount between image frames required for image processing at the time of moving image editing or the like. The present invention relates to a method for compensating a luminance change of a moving image to be compensated, a moving image encoding device, a moving image decoding device, a recording medium on which a moving image encoding or decoding program is recorded, and a recording medium on which encoded data of the moving image is recorded.

[0002]

2. Description of the Related Art As a coding technique for efficiently transmitting / accumulating moving picture data, a technique called motion compensation inter-frame predictive coding has been widely known. This technology does not transmit / accumulate the luminance (or color difference) of the processing target image frame itself, but instead converts the already encoded moving image frame (referred to as a reference image frame) to the processing target image frame such as an object in a screen. Is detected, and the luminance difference value between the positions shifted by the movement is transmitted / stored. Since the difference value is generally smaller than the luminance itself, it is possible to transmit / accumulate a small amount of data by using the motion-compensated inter-frame prediction coding.

Here, when detecting a motion, it is generally assumed that the illumination conditions and the like are the same between the image frame to be processed and the reference image frame. That is,
The motion is detected on the assumption that the corresponding objects actually have the same luminance value in both frames. However, for example, when flickering occurs, when a strobe illuminates, or by adjusting the aperture of a camera or using a video editing device, the entire screen becomes brighter (hereinafter referred to as “fade-in”) or darker (hereinafter referred to as “fade-out”). ")
In the case where the brightness itself changes with time, as in the case of such a case, the brightness values of the actually corresponding objects are greatly different from each other. As a result, even if the motion compensation inter-frame prediction coding is performed, there is a drawback that (1) motion detection is not performed correctly and (2) coding efficiency is not sufficiently improved.

[0004] Among the above-mentioned problems, as a prior art mainly for solving the problem (1), a technology for detecting a luminance change occurring between both image frames, and a method for correcting the detected luminance change. There is a technique for performing motion detection. Brightness changes can be roughly classified into two types: those having different degrees of change depending on the spatial position of the image frame, such as those caused by noise, and those having the same degree of change in the entire image frame, such as those caused by fade-in / fade-out. Kinds of things are conceivable. Here, in order to solve the problem (1), since it is important to detect the luminance change as accurately as possible, a small area of the frame is required so as to cope with either type of luminance change. It is often detected every time. However, this alone is problem (1)
Is solved, the inter-frame luminance difference value is still a large value, and the problem (2) cannot be solved. Therefore, there is a technique for transmitting / accumulating the detected luminance change for each small area as encoded data, and correcting the luminance change on the decoder side, thereby reducing the inter-frame luminance difference value. However, in this technique, since the encoded data representing the luminance change is transmitted for each small area, the data amount required for the small area eventually increases, and the encoded data amount as a whole is not so much, that is, the encoding efficiency does not improve much.

On the other hand, when the same luminance change occurs in the entire image frame, in order to improve the coding efficiency, only a single luminance change is detected in the entire image frame, and the luminance change is detected. There is a technique of transmitting / accumulating encoded data. In this technique, since only one encoded data representing a luminance change needs to be sent for each image frame, the amount of the encoded data is extremely small, and the inter-frame luminance difference value can be calculated only by correcting the luminance change. Since it can be reduced to some extent, it greatly contributes to improvement of coding efficiency in the case of fade-in / fade-out. As a parameter representing a general luminance change amount specifically mentioned in this technique, only a parameter indicating a gain change is used. That is, a parameter indicating a change in gain and D _B, the luminance value x of each pixel are assumed to be changed to x 'of the formula (1). x ′ = x + D _B (1) In this case, the luminance change amount Δx is Δx = x′−x = D _B (2) and is always constant.

FIG. 3 shows a case where still images containing various brightnesses are actually photographed by a camera, and the aperture of the camera is gradually reduced to darken the entire screen, that is, a case where the image fades out. Of the respective brightness values are displayed. For example, 1st frame and 11th frame, or 11
Looking at the change in each of the luminance values of the frames and the 21st frame, the change in luminance is substantially constant regardless of the value of the luminance value itself. Therefore, in this case, the assumption of Expression (1) is almost satisfied. That constant variation thereof is D _B in the formula (1).

[0007]

On the other hand, FIG. 4 shows the brightness values when the same still image is photographed and recorded for a while without changing the aperture of the camera, and later faded out by a digital editor. Is displayed.
For example, when looking at the change in each of the luminance values in the 41st and 51st frames or in the 51st and 61st frames, the change in luminance changes depending on the value of the luminance value itself. That is, when the luminance value itself is large, the amount of change in luminance is large, and as the luminance value itself is small, the amount of change in luminance is small. Therefore, this case is not composed premise of formula (1), resulting in accuracy becomes insufficient to represent the global luminance variation only parameter D _B. By the way, when an image is divided into small areas, for example, 8 × 8 pixels, variations in luminance in the small areas are generally not so large. For example, when the luminance value of the entire image frame changes from 0 to 255, in most cases, there is only a variation of about 20 to 40 in the small amount region. Within this range, the change in luminance can be regarded as substantially constant regardless of the value of the luminance value itself. That is, when the luminance change is obtained in the small area, it is almost sufficient even under the assumption of the expression (1). However, the range of the luminance value is wide in the entire image frame. That is, a luminance value of 0 to 255
Is highly likely to occur over the entire range. Therefore, when one type of luminance change is obtained for the entire image frame, the approximation of the expression (1) is insufficient.

SUMMARY OF THE INVENTION The present invention has been made in view of the above circumstances, and a method of compensating for a luminance change of a moving image, which allows more efficient encoding when encoding a moving image having a global luminance change. It is an object of the present invention to provide a moving image encoding device, a moving image decoding device, a recording medium recording a moving image encoding or decoding program, and a recording medium recording moving image encoded data.

[0009]

In order to achieve the above object, according to the first aspect of the present invention, there is provided an encoding processing for transmitting and storing a digital moving picture or an image processing for editing a moving picture. , A parameter representing a gain change which is a global luminance change amount between a processing target image plane and a reference image plane is denoted by D _B , a parameter indicating a contrast change is denoted by D _C , a luminance variation compensation method of a moving image, characterized in that to compensate for the global luminance variation by modifying the luminance value x of the pixel in D _C · x + D _B. The reference invention of claim 2, in the luminance variation compensation method of a moving image according to claim 1, the parameter D _B representing the gain change, the parameter D _C is representative of a contrast change, the processing target image plane It is characterized in that it is obtained on condition that the sum of squares of errors between corresponding pixels in the image plane is minimized.

[0010] Further, the invention described in claim 3 is based on claim 2.
N in the luminance variation compensation method of the moving image, wherein the parameter D _B representing the gain change, the parameter D _C is representative of a contrast change, the processing target image plane, consisting of K × L pixels (K, L is a natural number) in (n is an integer of 2 or more) is calculated two parameters D _C, D _B for each block of the most frequent higher parameter D _C, the set of D _B, parameter D _C of the processing target image plane, D It is characterized by deciding as _B. According to a fourth aspect of the present invention, in the moving image luminance change compensating method according to the second aspect, the parameter D _B representing the gain change and the parameter D _C representing the contrast change are the same as those of the image plane to be processed. N (n, n) consisting of K × L pixels (K and L are natural numbers)
Is an integer of 2 or more), two parameters D _C , D _B are calculated for each block, and a set of average values of the calculated parameters D _C , D _B is converted into the parameters D _C , D _C ,
It is characterized by determining the D _B. Claim 5
The invention according to claim 1, wherein in the moving image luminance change compensation method according to claim 1, the image plane includes an image frame,
Alternatively, it is characterized by being a video object plane (VOP).

[0011] Next, an invention according to claim 6, parameter D _B representing the gain change, the parameters D _C representing a contrast change, represents the global brightness variation between the processing target image plane and the reference image plane features and parameter determining means for determining a parameter, that the luminance value x of each pixel of the reference image planes, the parameter D _B, using D _C, and a luminance variation compensation means for correcting the D _C · x + D _B Is a moving image encoding device. Further, an invention according to claim 7, in the moving picture coding apparatus according to claim 6, wherein the parameter determining means, the parameter D _B, a D _C, corresponding with the reference image plane and the target image plane It is characterized in that it is obtained on the condition that the sum of squares of the error between pixels is minimized.

[0012] The invention described in claim 8 is the same as in claim 7.
In the moving picture encoding device described in the above, the parameter determination means may include a K × L pixel (K,
L is a natural number, and two parameters D _C and D _B are calculated for each of n (n is an integer of 2 or more) blocks, and a set of the most frequent parameters D _C and D _B is processed as an image to be processed. It is characterized in that it is determined as plane parameters D _C and D _B. According to a ninth aspect of the present invention, in the moving picture coding apparatus according to the seventh aspect, the parameter determining means includes n × K × L pixels (K and L are natural numbers) of the processing target image plane. (N is an integer of 2 or more)
, Two parameters D _C and D _B are calculated for each block, and a set of average values of the calculated parameters D _C and D _B is determined as the parameters D _C and D _B of the processing target image plane. . The invention according to claim 10 is
7. The video encoding device according to claim 6, wherein the image plane is an image frame or a video object plane (VOP).

[0013] Next, the invention of claim 11, the parameter D _B representing the gain change is global brightness variation between a process target image plane obtained reference image plane in the encoding process of a moving picture moving image and by using the parameters D _C representing a contrast change, characterized in that the luminance value x of each pixel of the reproduced image plane decoding is finished with a luminance variation compensation means for correcting the D _C · x + D _B It is a decoding device. According to a twelfth aspect of the present invention, in the moving picture decoding apparatus according to the eleventh aspect, the image plane is an image frame or a video object plane (VOP).

Next, according to a thirteenth aspect of the present invention, a moving image
Computer readable recording encoding program
Parameter that represents the gain change
D_B, A parameter D representing a change in contrast_CThe processing
Global image between the elephant image plane and the reference image plane
Parameter determination to be used as a parameter representing the amount of change in brightness
A luminance value x of each pixel of the reference image plane.
With the parameter D_B, D_CUsing D_C・ X + D_BOsamu
And causing the computer to execute the correct luminance change compensating means.
Recording medium for recording a moving picture coding program for
You. The invention described in claim 14 is the same as the invention described in claim 13.
On the recording medium on which the video coding program
The parameter determining means determines that the parameter D _B,
D_CBetween the image plane to be processed and the reference image plane
Condition that the sum of squares of the corresponding pixel error is minimized
It is characterized by seeking.

According to a fifteenth aspect of the present invention, in the recording medium storing the moving picture coding program according to the fourteenth aspect, the parameter determining means determines that the number of K × L pixels (K , L is a natural number)
(N is an integer of 2 or more) is calculated two parameters D _C, D _B for each block of the most frequent higher parameter D _C, the set of D _B, parameter D _C of the processing target image plane, D It is characterized by deciding as _B. According to a sixteenth aspect of the present invention, in the recording medium on which the moving image encoding program according to the fourteenth aspect is recorded, the parameter determining means determines that a K × L pixel (K, L two parameters D _C for each block of n (n is an integer of 2 or more) composed of natural numbers), calculates D _B, calculated parameter D _C, a set of the average value of D _B,
Parameter D _C of the processing target image plane, is characterized by determining the D _B. According to a seventeenth aspect of the present invention, in the recording medium storing the moving picture encoding program according to the thirteenth aspect, the image plane is an image frame or a video object plane (VO).
P).

Next, an invention according to claim 18 is a computer-readable recording medium on which a moving picture decoding program is recorded, wherein the processing target image plane and the reference picture plane obtained in the moving picture encoding step are obtained. by using the parameters D _C representing the parameters D _B and contrast change representing the gain change is global brightness variation between the luminance value x of each pixel of the reproduced image plane decoding is finished D _C · x + D _B This is a recording medium on which a moving image decoding program for causing a computer to execute a luminance change compensating means for correcting the moving image is recorded. According to a nineteenth aspect of the present invention, in the recording medium recording the moving image decoding program according to the eighteenth aspect, the image plane is an image frame or a video object plane (VOP).
It is characterized by being.

According to a twentieth aspect of the present invention, there is provided a computer-readable recording medium recording encoded data of a moving image, wherein the encoded data is an image plane encoded in units of image planes. includes a plurality of areas for recording encoded data, area for recording said image plane encoding data, parameter D _B representing the gain change is global brightness variation between the processing target image plane reference image plane, and a parameter area for recording the parameters D _C representing a contrast change, the luminance value x of each pixel of the reference image planes, the parameter D
_B, with a D _C, and an image plane which is modified D _C · x + D _B, encoded recorded by encoding data relating to the difference value of the pixel values of the corresponding each pixel in the image plane units in said processing object image plane This is a recording medium on which encoded data of a moving image including an area is recorded. Claim 2
According to a first aspect of the present invention, in the recording medium recording the encoded data of the moving image according to the twentieth aspect, the image plane is an image frame or a video object plane (VOP).

[0018]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The outline of a moving picture luminance change compensation method, a moving picture coding apparatus, and a moving picture decoding apparatus according to the present invention will be described below. Note that luminance change compensation of a moving image and encoding and decoding of the moving image using the luminance change compensation are performed in units of image planes. Here, the “image plane” is an “image frame” corresponding to one frame of a moving image, or a “video object plane” (Video Object Plain) which is an image corresponding to a temporal change of each content of the moving image.
ne: hereinafter abbreviated as “VOP”).
In the following, description will be given by taking an image frame among image planes as an example. The VOP will be described separately in detail.

First, a method for compensating for a luminance change of a moving image according to the present invention will be described. Luminance variation compensation method of a moving image of the present invention, a parameter representing the gain change D _B, when a parameter representing a contrast change was D _C, by modifying the luminance value x of each pixel in D _C · x + D _B Compensate for global luminance change. Luminance variation compensation method of a moving image of the present invention, in order to solve the above problems, the luminance value x of each pixel, the parameters D _C showing parameters D _B and the contrast changes indicative of gain change, Equation (3) Is assumed to change to x ′. _{x '= D C · x +} D B ····· (3) In this case, the luminance variation △ x is △ x = x'-x = ( D C -1) · x + D B ····· (4) Which depends on the luminance value x of the pixel. As described above, when one type of luminance change is obtained for the entire image frame, using the expression (3) that can express the luminance change depending on the value of the luminance value itself requires the use of a fade-in / fade-out method for the entire screen. This has great significance in improving the coding efficiency when a luminance change occurs.

According to an embodiment of the present invention, a parameter D _B representing the gain change, the parameters D _C representing a contrast change, the square sum of errors of the corresponding pixels between the processing target image frame and the reference image frame Min It is sought on condition that it becomes. The luminance variation detection method of the moving image of the present invention, first, the two parameters D _B, the processed image frames _{D C K × L (K,} L is a natural number) detected for each block of pixels. At this time, a criterion for calculating the parameters is that the sum of the squares of the errors between the corresponding pixels in the processing target image frame and the reference image frame be minimized. Next, 2 is calculated for each block of K × L pixels.
Types of parameters D _B, determines the value of the average value or the maximum frequency of each of the D _C as a parameter to process the entire target frame.

In the following, a calculation formula for detecting parameters for each block of K × L pixels will be derived. Now, i th two in the block of parameters each _{D B (i), D C} (i)
The pixel value of the block of K × L pixels in the frame to be processed is y _{k, l} , and the pixel value of the block of K × L pixels in the reference image frame is x _{k, l} . As described above, as a criterion for parameter calculation, the square sum J of the error between the corresponding pixels in the processing target image frame and the reference image frame is used.
Is minimized. That is,

(Equation 1) And J is partially differentiated with D _C (i) and D _B (i), respectively, and set to 0.

(Equation 2) From equations (6) and (7)

(Equation 3)

However,

(Equation 4) It is.

As described above, the parameters D _C (i) and D _B (i) calculated for each block of K × L pixels by the equations (8) and (9) are the same as those of the K × L pixels. the value of the average value or the maximum frequency of the two types of parameters calculated for each block, determining for the entire frame to be processed parameter D _C, as D _B. Moving picture coding apparatus of the present invention, parameter determination for determining parameters D _B representing the gain change, the parameters D _C representing a contrast change, as a parameter representing the global luminance variation between the reference image frame processed image frame Means, and a luminance value x of each pixel of the reference image frame,
_B, with a D _C, has a luminance variation compensation means for correcting the D _C · x + D _B.

According to an embodiment of the invention, the parameter determining means determines the parameter D _B, a D _C, the sum of squares of errors of the corresponding pixels between the processing target image frame and the reference image frame is minimized Is required. According to an embodiment of the present invention, the parameter determination unit determines that each of n (n is an integer equal to or greater than 2) blocks of K × L pixels (K and L are natural numbers) of the image frame to be processed.
One of the parameters D _C, calculates D _B, most frequent higher parameter D _C, the set of D _B, parameter D _C of the processed image frame is determined as D _B. According to an embodiment of the present invention, the parameter determining means performs two (n is an integer of 2 or more) blocks of K × L pixels (K and L are natural numbers) of the image frame to be processed. The parameters D _C and D _B are calculated, and the calculated parameters D _C and D
A set of average values of _B, the parameters D _C of the processed image frame is determined as D _B. In the video decoding apparatus of the present invention, the parameter D _B obtained in the moving picture coding apparatus, using a D _C, correct the luminance value x of each image of the reproduced image frame decoding has ended D _C · x + D _B Brightness change compensating means.

Next, an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing a moving picture coding apparatus using the moving picture luminance change compensation method of the present invention. The main moving picture coding apparatus comprises a human terminal 1, frame memories 2 and 3, a motion detector 4, a luminance change detector 5, a frequency calculator 6, a luminance change compensator 7, an output terminal 8, and a frame memory 9.
, A motion detection / compensation unit 10, a subtractor 11, a discrete cosine transform unit 12, a quantization unit 13, an output terminal 14, and an inverse quantization unit 1.
5, an inverse discrete cosine transform unit 16, an adder 17, and an output terminal 18. An N-th frame original image is input from an input terminal 1 to a frame memory 2, a frame memory 3, a motion detector 4, and a luminance change detector 5. Frame memory 3
Then, the Nth frame original image is stored, and the (N-1) th frame original image stored so far is sent to the motion detecting section 4 and the luminance change detecting section 5.

In the motion detecting section 4, as shown in FIG.
The motion between the (N-1) th frame original image and the (N-1) th frame original image is obtained for each block of 16 × 16 pixels in the frame original image by the block matching method. The number of all blocks in one frame is n, and each block is represented by B1, B2,.
i,..., Bn. Also, the motion vector calculated for the block Bi is Vi. Motion vector V
i is sent to the luminance change detecting unit 5. Brightness change detector 5
Then, in the block Bi of the Nth frame original image, 16
Each of the luminance values of × 16 pixels and 16 × pixels located at a position shifted by the motion vector Vi in the (N−1) th frame original image
Using each luminance value of 16 pixels, Expression (1) is used to calculate Expression (1).
0), two parameters D in block Bi
C (i) and DB (i) are calculated. The obtained two parameters D _C (i) and D _B (i) are sent to the frequency calculation unit 6. The frequency calculation section 6, D _C (i), calculated for one frame the frequency of D _B (i), a set of most frequently higher parameter first N
The parameters are determined as parameters D _C and D _B representing the global luminance change between the frame original image and the (N−1) th frame original image, and the values are sent to the luminance change compensator 7 and the output terminal 8 is output. Output to outside.

On the other hand, the Nth frame original image input from the input terminal 1 is delayed by one frame in the frame memory 2 and then input to the subtractor 11 manually. The (N-1) th frame reproduced image is stored in the frame memory 9, and the luminance change compensating unit 7 calculates the expression (3) for each luminance value x of the image.
Is used to perform luminance change compensation, and the luminance change compensated value x '
Is input to the motion detection / compensation unit 10. The motion detection / compensation unit 10 performs motion detection and motion compensation by block matching for each block of 16 × 16 pixels.
The motion-compensated value is input to the subtractor 11 and the adder 17. The subtractor 11 takes the difference between the two data and sends it to the discrete cosine transform unit 12. In the discrete cosine transform unit 12, 16 ×
One block of data consisting of 16 pixels is subjected to discrete cosine transform in units of 8 × 8 pixels, and the obtained discrete cosine transform coefficients are quantized by the quantization unit 13. The quantized discrete cosine transform coefficients are sent out from the output terminal 14, are inversely quantized by the inverse quantizer 15, and are further inverse discrete cosine transformed by the inverse discrete cosine transform unit 16. Adder 17
In, the data sent from the motion detection / compensation unit 10 is added to the data, and the data is stored in the frame memory 9 as an N-th frame reproduced image. The Nth frame reproduced image stored in the frame memory 9 is used as a reference image when decoding the image of the next frame (the (N + 1) th frame). Note that the parameters D _C (i) and D _B (i)
The respective average values of (i = 1 to n) may be determined as the parameters D _C and D _B.

The data encoded by the present moving picture coding apparatus is decoded by the moving picture decoding apparatus shown in FIG. 2 using the moving picture luminance change compensation method of the present invention. The main video decoding device includes an input terminal 21-23 and an inverse quantization unit 24.
And an inverse discrete cosine transform unit 25, an adder 26, and an output terminal 2.
7, a frame memory 28, a luminance change compensator 29 and a motion compensator 30. The quantized discrete cosine transform coefficient output from the output terminal 14 of FIG. The input terminal 22 includes the output terminal 1 shown in FIG.
The motion vector for each block output from 8 is input. To the input terminal 23, two parameters D _C and D _B representing the luminance change amount for each frame output from the output terminal 8 of FIG.

Now, the block Bi (1) of the N-th frame image
It is assumed that encoded data of (6 × 16 pixels) is input.
That is, the quantized discrete cosine transform coefficient in the block Bi is input to the input terminal 21. The input terminal 22 receives the motion vector Vi of the block Bi. Note that two parameters D _C and D _{B that} are values common to the entire N-th frame have already been input to the input terminal 23.
The quantized discrete cosine transform coefficient manually input from the input terminal 21 is inversely quantized by the inverse quantizer 24, inversely discrete cosine transformed by the inverse discrete cosine transform unit 25, and added to the adder 26.
Sent to

The (N-1) th frame reproduced image data which has already been decoded is stored in the frame memory 28, and the luminance change compensator 29 calculates the expression (3) for each luminance value x of the image. , And the value x ′ subjected to the luminance change compensation is input to the motion compensation unit 30. In the motion compensating unit 30, a value whose motion is compensated by the motion vector Vi is input to the adder 26. The adder 26 obtains an N-th frame reproduced image by adding the data from the inverse discrete cosine transform unit 25 and the data from the motion compensating unit 30, outputs the N-th frame reproduced image from the output terminal 27, and stores it in the frame memory 28. Each of them is used as a reference image when decoding the image of the next frame (the (n + 1) th frame).

Next, the moving picture coding apparatus shown in FIG.
Computer-readable encoded data generated by
The structure of data when recording on a simple recording medium will be described with reference to FIG.
explain. Here, the encoded data simply represents an image frame.
Encoded image frames encoded as digits
It consists of. In FIG. 6, each image frame
Has the same structure of encoded data, but as an example,
N-th frame, which is encoded data of the image frame of the eye
This will be described with reference to the encoded data of FIG. As shown in FIG.
In addition, the encoded data of the Nth frame includes a parameter D_C・
81, parameter D_B.Parameter which is an area for recording 82
The meter area 70 and the luminance value of each pixel of the reference image frame
x with this parameter D_B・ 81, D_CD using 82
_C・ X + D_BAnd the Nth image frame
Difference data of the pixel value of each corresponding pixel in the frame
Encoding area 71 in which data relating to
It consists of. Furthermore, the coding area 71
The encoded data for each block of 16 × 16 pixels shown in FIG.
It is composed of the stored block areas 72 and 73. What
In addition, as shown in FIG.₁
~ B_nIf it is up to the coding area as shown in FIG.
71 is composed of n block areas.
You.

Each block area is composed of a motion vector V _{i in} block units and a block of 4 × 8 × 8 pixels.
It is divided into four sub-blocks, and is constituted by _four areas C _{i-1 to} C _i-4 for storing data encoded for each sub-block. Taking the first block area 72 in FIG. 6 as an example, reference numeral 91 indicates a motion vector, and reference numerals 92 to 9
5 corresponds to the data encoded for each sub-block.
Here, the encoded data of FIG. 6, will be described the correspondence between the output from the video encoding apparatus of FIG. 1, the parameter D _C ·
81, D _B · 82, the output terminal 8 of the frequency calculation unit 6 of FIG. 1
Corresponding to the output value from. Then, the motion vector (for example, reference numeral 91) in each block region corresponds to the output value from the output terminal 18 of the motion detection / compensation unit 10 in FIG.
Data encoded for each sub-block (for example, code 92
To 95) correspond to the output value from the output terminal 14 of the quantization unit 13.

In the case of the encoded data structure shown in FIG. 6, the output operation of the moving picture encoding apparatus shown in FIG. 1 is as follows. First, and outputs the parameter _{D C · 81, D B ·} 82 indicating the luminance change of the entire image frame. Next, FIG.
And the motion vector V1 · 91 of the block B1 shown in the sub-block SB _1-1 to SB every _1-4 encoded data C _1-1 · 92 to
C _1-4 · 95 is output. Then, the motion vector V _i for each block, sub-block SB _i-1 ~SB _i-4 of the coded data C _i-1 ~C _i-4 are sequentially output, performs similarly to the block Bn. This operation is performed for each image frame, and encoded data is recorded.

The decoding of the encoded data shown in FIG.
Parameter D _C · 81 parameter region of FIG. 6 to the input terminal 23 of the video decoding apparatus shown in FIG. 2, D _B · 82 enter the input terminal 2 a motion vector Vi of each block region
2, decoding is performed for each image frame by inputting the coded data C _{i-1 to} C _i-4 of the sub-block to the input terminal 21.

FIG. 7 shows a parameter D for each block together with a motion vector 55 and coded data 56 to 59 for each sub-block for each block area denoted by reference numerals 51 and 52.
Illustrates a structure of encoded data for recording _{C · 53, D B · 54} . Note that when fade-in / fade-out, even when obtained for each block parameter D _C, the D _B, 1 single parameter D _C for the image frame, even when looking for D _B, code in the sub-block There is no significant difference in the conversion efficiency. Here, for example, the size of the image frame is 352 × 240 pixels (the number of blocks of 16 × 16 pixels:
At 330), if the number of bits of each of the parameters D _C and D _B is 8 bits, the encoded data shown in FIG. 6 is 5264 bits per image frame (= {8 + 8} × {330-1}) can be reduced.

[0036] Figure 8 records only the parameter D _C · 61 representing one gain change for each image frame, reference numeral 62
FIG. 7 is a diagram showing the structure of encoded data when the configuration of the block area shown in FIG. Here, assuming that the number of bits of the parameters D _C and D _B is 8 bits, the number of bits in the parameter area is 8 per image frame in the encoded data in FIG. 8 more than in the encoded data in FIG.
A bit less. However, in the fade-in / fade-out, when the luminance variation is changed by the value of the luminance value itself as shown in FIG. 4, the luminance value x of each pixel of the reference image planes, and modified using the parameter D _C Also, the approximation of the luminance change is insufficient. Therefore, the difference value subjected to the discrete cosine transform becomes a relatively large value, and as a result, the encoded data of each sub-block (for example, codes 64 to
67) increases, and the amount of encoded data increases when viewed in one entire image frame. As described above, the coding efficiency can be increased by configuring the encoded data as shown in FIG.

In the above-described embodiment, an example has been described in which the parameters D _C and D _B are output in units of image frames. However, the present invention is not limited to this, and units corresponding to image frames, for example, standardization of color moving images MPEG-4 (MovingPicture Experts Group)
-Video object plane (V) in Phase 4)
OP) may be used as a unit. That is, the parameters D _C and D _B may be obtained between the reference VOP and the processing target VOP, and the compensation of the luminance change and the encoding / decoding using these parameters may be performed. The configuration of the encoded data in this case is such that, in the encoded data configuration of MPEG-4, the parameters D _C and D _B and the encoded data of the VOP whose luminance change has been compensated using these parameters are recorded as a pair. Will be. That is, when the data structure of the MPEG-4, when or contain fixed data header at the head of a frame, for example, discrete cosine transform result of the sub-block SB _1-1 of Figure 5 are all zero The coded data _C1-1 in FIG. 6 differs from the coded data for each frame shown in FIG. 6 in that the coded data C1-1 in FIG. There is a point. Although there is such a difference, in FIG. 6, the parameters D _C and D _B of the parameter area 70 and the area of the coded data of the VOP whose luminance change has been compensated using these parameters (the coding area 71 of FIG. The same) is recorded in pairs.

For details of the structure of standard encoded data of MPEG, see T. Ebrahimi, "MPEG-4 video verificati
on model: A video encoding / decodingalgorithm based
on content representation ", Signal Processing: Im
age Communication vol.9, no.4, pp.367-384, May 199
It is described in 7.

Here, the hierarchical structure in MPEG-4 will be briefly described with reference to the drawings. FIG. 9 is a diagram showing an example of a hierarchical structure of moving image data in MPEG-4.
That is, the entire moving image is composed of a group corresponding to one shot called one or more video objects (hereinafter abbreviated as “VO”). FIG. 9 shows two shots VO1 and VO2. Then, the VO has one or more video object layers (Vide
o Object Layer: a collection of moving image contents called "VOL". For example, in FIG. 9, VOL1 is a snowman VPL11 and a house VOL12.
Consists of a set of two VOLs, and VO2 is a square VPL2
It consists of a set of three VOLs: 1, triangular VOL22 and XVOL23.

A VOL is composed of one or more video object planes, each of which is a group of images whose contents change with time. For example, VOL11 is composed of a group of three VOPs VOP1 to VOP3. Then, the VOP may be encoded using the parameters D _C and D _B as described above.

In the above embodiment, the discrete cosine transform has been described as an example in encoding and decoding. However, the present invention is not limited to this, and other orthogonal transforms such as Hadamard transform and wavelet transform are used. May be used. Further, in the above embodiment, the case where the block is 16 × 16 pixels and the sub-block is 8 × 8 pixels as shown in FIG. 5 has been described, but the present invention is not limited to this.
Also, a program for realizing the operation of the moving picture encoding device shown in FIG. 1 or the moving picture decoding device shown in FIG. 2 is recorded on a computer-readable recording medium,
The program recorded on this recording medium may be read and executed by a computer to encode or decode a moving image.

[0042]

As described above, according to the present invention, there is provided a method for compensating a luminance change of a moving picture, a moving picture coding apparatus, a moving picture decoding apparatus, a recording medium on which a moving picture coding or decoding program is recorded, and a moving picture code. According to the recording medium on which the encrypted data is recorded, the following effects can be obtained. According to the present invention, when detecting or compensating for a global luminance change, by using a parameter representing a change in contrast in addition to a parameter representing a change in gain, a moving image having a global luminance change is used. At the time of encoding, there is an effect that encoding can be performed with higher efficiency than before.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a moving picture encoding apparatus using a moving picture luminance change compensation method for a moving picture of the present invention.

FIG. 2 is a block diagram showing a moving picture decoding apparatus using the moving picture luminance change compensation method of the present invention.

FIG. 3 is a diagram for explaining a global change in luminance of an image.

FIG. 4 is a diagram for explaining a global change in luminance of an image.

FIG. 5 is a diagram for explaining how to divide an image frame.

FIG. 6 is a diagram illustrating a configuration of encoded data according to the present invention.

FIG. 7 is a diagram showing a configuration of encoded data when an area for recording a parameter is provided for each block.

8 is a diagram showing a configuration of the encoded data obtained when a region for recording only the parameter D _B representing the gain change for each image frame.

FIG. 9 is a diagram for describing a video object (VO), a video object layer (VOL), and a video object plane (VOP) in MPEG-4.

[Explanation of symbols]

Reference Signs List 1 input terminal 2, 3 frame memory 4 motion detector 5 luminance change detector 6 frequency calculator 7 luminance change compensator 8 output terminal 9 frame memory 10 motion detector / compensator 11 subtractor 12 discrete cosine transform unit 13 quantization Unit 14 output terminal 15 inverse quantization unit 16 inverse discrete cosine transform unit 17 adder 18 output terminal 21 to 23 input terminal 24 inverse quantization unit 25 inverse discrete cosine transform unit 26 adder 27 output terminal 28 frame memory 29 brightness change compensation Unit 30 motion compensation unit

Claims (21)

[Claims] 1. A method for compensating a luminance change of a moving image in an encoding process for transmitting and storing a digital moving image or in an image process at the time of editing a moving image, wherein a global change between a processing target image plane and a reference image plane is provided. parameter D _B representing the gain change is a luminance change amount, a parameter representing a contrast change and D _C, to compensate for the global luminance variation by modifying the luminance value x of each pixel in D _C · x + D _B A method for compensating luminance change of a moving image, characterized in that 2. A parameter D _B representing the gain change,
Parameter D _C representing the contrast change, the moving picture according to claim 1, the sum of squares of errors between the pixels corresponding processing target image plane and the reference image plane and finding in the condition that the minimum Brightness change compensation method. 3. A parameter D _B representing the gain change,
The parameter D _C representing the change in contrast is two parameters D _C and D _B for each of n (n is an integer of 2 or more) blocks of K × L pixels (K and L are natural numbers) of the image plane to be processed. calculating the most frequent higher parameter D _C, the set of D _B, parameter D _C of the processing target image plane, the luminance variation compensation method according to claim 2, wherein determining that D _B. 4. A parameter D representing the gain change._B,
Parameter D representing contrast change_CIs the K × L pixel of the processing target image plane (K and L are natural
Number) (n is an integer of 2 or more)
About two parameters D_C, D_BIs calculated, and the calculated
Parameter D_C, D_BThe set of average values of
Parameter D _C, D_BThe contract is characterized by
A method for compensating for a change in luminance according to claim 2. 5. The method according to claim 1, wherein the image plane is an image frame or a video object plane (VOP). 6. The parameter D _B representing the gain change, the parameters D _C representing a contrast change, and parameter determining means for determining a parameter representing a global brightness variation between the processing target image plane reference image plane, the reference the luminance value x of each pixel of the image plane, the parameter D _B, using D _C, the video encoding apparatus characterized by comprising a luminance variation compensation means for correcting the D _C · x + D _B. 7. The parameter determining means obtains the parameters D _B and D _C on condition that the sum of squares of errors between corresponding pixels in the processing target image plane and the reference image plane is minimized. Claim 6
5. The video encoding device according to item 1. 8. The parameter determining means includes two parameters D _C , N for each block of n × K pixels (K and L are natural numbers) of an image plane to be processed, where n is an integer of 2 or more. calculating a D _B, most frequent high was parameter D _C, the set of D _B, parameter D _C of the processing target image plane, the moving picture coding according to claim 7, wherein determining that D _B apparatus. 9. The image processing apparatus according to claim 1, wherein the parameter determining unit is configured to determine a number of K × L pixels (K and L are natural
Number) (n is an integer of 2 or more)
About two parameters D_C, D_BIs calculated, and the calculated
Parameter D_C, D_BThe set of average values of
Parameter D _C, D_BThe contract is characterized by
A moving picture coding apparatus according to claim 7. 10. The moving picture coding apparatus according to claim 6, wherein said picture plane is a picture frame or a video object plane (VOP). 11. the parameters D _C representing the parameters D _B and contrast change representing the gain change is global brightness variation between the obtained processed image plane and the reference image plane in the encoding process of a moving picture used, video decoding apparatus, wherein a luminance value x of each pixel of the reproduced image plane decoding is finished with a luminance variation compensation means for correcting the D _C · x + D _B. 12. The moving picture decoding apparatus according to claim 11, wherein the picture plane is a picture frame or a video object plane (VOP). 13. A computer-readable recording medium recording a moving picture encoding program, the parameters D _B representing the gain change, the parameters D _C representing a contrast change, between the reference image plane and the target image plane and parameter determining means for determining a parameter representing the global luminance variation, the luminance value x of each pixel of the reference image planes, the parameter D _B, using D _C, the luminance variation compensation to correct the D _C · x + D _B Recording medium for recording a moving image encoding program for causing a computer to execute the means. 14. The parameter determining means obtains the parameters D _B and D _C on condition that the sum of squares of errors between corresponding pixels in a processing target image plane and a reference image plane is minimized. Claim 1
4. A recording medium on which the moving picture encoding program according to 3 is recorded. 15. The parameter determining means comprises two parameters D _C , n for each of n (n is an integer of 2 or more) blocks of K × L pixels (K and L are natural numbers) of an image plane to be processed. calculating a D _B, most frequent high was parameter D _C, the set of D _B, parameter D _C of the processing target image plane, the moving picture coding according to claim 14, wherein determining that D _B A recording medium on which a program is recorded. 16. The apparatus according to claim 1, wherein the parameter determining unit is configured to calculate K × L pixels (K and L are natural
Number) (n is an integer of 2 or more)
About two parameters D_C, D_BIs calculated, and the calculated
Parameter D_C, D_BThe set of average values of
Parameter D _C, D_BThe contract is characterized by
A program which records the moving picture coding program according to claim 14.
Recording medium. 17. The recording medium according to claim 13, wherein the image plane is an image frame or a video object plane (VOP). 18. A computer-readable recording medium on which a moving image decoding program is recorded, wherein the global luminance change amount between a processing target image plane and a reference image plane obtained in a moving image encoding step. by using the parameters D _C representing the parameters D _B and contrast change representing the gain change, the luminance value x of each pixel of the reproduced image plane decoding is finished D _C · x + D _B
A recording medium which stores a moving image decoding program for causing a computer to execute a luminance change compensating means for correcting the change. 19. The recording medium according to claim 18, wherein the image plane is an image frame or a video object plane (VOP). 20. A computer-readable recording medium on which encoded data of a moving image is recorded, wherein the encoded data includes a plurality of areas in which image plane encoded data encoded in units of image planes is recorded. The area in which the image plane coded data is recorded is a parameter D representing a gain change which is a global luminance change amount between the processing target image plane and the reference image plane.
_B, and a parameter area for recording the parameters D _C representing a contrast change, the luminance value x of each pixel of the reference image planes, the parameter D _B, using D _C, the image plane was modified to D _C · x + D _B And a coded area in which data relating to a pixel value difference value for each corresponding pixel in the processing target image plane is coded and recorded in units of image planes. 21. The recording medium according to claim 20, wherein the image plane is an image frame or a video object plane (VOP).