JP2629997B2 - Motion compensated interframe coding device - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a motion compensation interframe coding apparatus for a television signal.

2. Description of the Related Art In recent years, with the development of video coding technology, a motion-compensated prediction interframe coding device has been used as a high-efficiency color video coding device used in videophones, video conference systems, CD-ROMs, digital VTRs, and the like. Is being developed. For example, Norihiko Fuukiki, “Multidimensional Signal Processing of TV Images” (November 1988
A motion-compensated prediction interframe coding apparatus described in March 7, published by Nikkan Kogyo Shimbun, Chapter 7, High Efficiency Coding, pp213-pp291) is known.

The motion-compensated prediction inter-frame coding method can realize high coding efficiency, but has the following problems.

Whether or not to perform motion compensation is determined by: 1) the difference between the pixel value of the coded block and the predicted pixel value of the coded block based on the motion compensation. 2 the difference between the pixel value of the coded block and the pixel value of the previous frame. It is determined using

Therefore, in the portion where there are a large number of blocks for which determination is performed near the boundary of whether or not to perform motion compensation, blocks to be motion-compensated and blocks not to be motion-compensated are mixed. In particular, during parallel movement (panning) of the entire screen having a size equal to or larger than a certain value, if there is a block that is not motion compensated surrounded by motion compensated blocks in a flat portion of the screen, image quality degradation may be caused. Become.

Hereinafter, a conventional motion compensation interframe coding apparatus will be described with reference to FIG. In FIG. 3, reference numeral 41 denotes an input terminal to which an input television signal is input, and reference numeral 43 denotes a motion vector of an encoded block by comparing the image signal of the encoded block of the current frame with the reproduced image signal of the previous frame. A motion vector calculating unit, 44 is an image memory unit for storing the reproduced image signals of the current frame and the previous frame, 48 is a motion compensating unit for performing motion compensation on the reproduced image signal of the previous frame, and 50 is 2 for the motion compensated signal. An intra-loop filter section for performing a dimensional low-pass filter processing; a prediction error calculation section for performing a difference operation between an original image signal and a prediction signal of an encoded block to calculate a prediction error;
Is an orthogonal transformation unit for orthogonally transforming the prediction error signal, 56 is a quantization unit for quantizing the orthogonal transformation coefficient, 59 is an inverse orthogonal transformation unit for inversely orthogonally transforming the quantized orthogonal transformation coefficient, and 61 is a reproduced image of the current frame. , 63 is a prediction error coding unit that performs communication coding of a prediction error, 65 is a motion vector coding unit that performs communication channel coding of a motion vector, and 67 is a transmission frame using a prediction code and a motion vector code. A code memory unit for temporarily storing a transmission frame;
An output terminal 71 outputs a transmission signal.

The operation of the above-described configuration will be described. A television signal converted into a digital signal by an analog-to-digital converter (not shown) and divided into blocks of N pixels in the horizontal direction and N lines in the vertical direction is
An input television signal is input from an input terminal 41. The motion vector calculation unit 43 compares the input television signal 42 with the reproduced television signal 45 of the previous frame stored in the image memory unit 44, calculates the motion of the coding block as a motion vector, and calculates a motion vector signal 46. Output to At the same time, the motion vector calculation unit 43 uses the evaluation value at the time of calculation of the motion vector to determine whether the motion compensation is valid or invalid for the encoded block, and outputs the result to the motion vector signal 46 as a motion compensation control signal. Therefore, the motion vector and the motion compensation signal are superimposed on the motion vector signal 46. The motion compensator 48 (1) performs motion compensation on the reproduced television signal 45 of the previous frame with a motion vector when the motion compensation control signal indicates that the motion compensation is effective. When the motion compensation is invalidated, the reproduced television signal 45 of the previous frame is output as it is as the motion compensation signal 49.

The in-loop filter unit 50 performs a two-dimensional low-pass filter process on the encoded block motion-compensated using the motion vector, and outputs the result as a prediction signal 51.

The prediction error calculation unit 52 performs a difference operation between the input television signal 42 of the coding block and the prediction signal 51, and outputs the result as a prediction error signal 53. The orthogonal transform unit 54 performs an orthogonal transform on the prediction error signal 53, and outputs the prediction error signal
The prediction error orthogonal transform coefficient 55 is calculated by removing the high correlation between the 53 neighboring pixels. In many cases, as the orthogonal transform method, a discrete cosine transform that has high conversion efficiency and is feasible in terms of hardware is used. The quantization unit 56 is a quantizer whose quantization characteristic changes according to the generated code amount 57, and quantizes the prediction error orthogonal transform coefficient 55 to calculate a prediction error orthogonal transform quantization coefficient 58. Inverse orthogonal transform unit 59
Performs an inverse orthogonal transform of the prediction error orthogonal transform quantization coefficient 58, and calculates a prediction error signal 60 including a quantization error. The reproduction image calculation unit 61 adds the prediction error signal 60 including the quantization error and the prediction signal 51 to calculate a reproduction image 62 of the encoded block. The image memory 44 stores the reproduced image signal 62 of the current frame and outputs the reproduced image signal 45 of the previous frame. The prediction error encoding unit 63 encodes the prediction error orthogonal transform quantization coefficient 58 and calculates a prediction error code 64. The motion vector encoding 65 encodes the motion vector 46 to calculate a motion vector code 66. The multiplexer 67 calculates a transmission frame 68 of a predetermined format from the prediction error code 64 and the motion vector code 66.
The code memory unit 69 temporarily stores the transmission frame 68, and synchronizes with a clock signal input from outside (not shown),
It is output from the output terminal 71 as a transmission code 70.

However, in the above configuration, whether or not to perform motion compensation is determined based on a prediction error signal based on motion compensation of a coded block and a difference value between the coded block and a previous frame. Therefore, according to these two values, in a portion where many blocks are determined near the boundary of whether or not to perform motion compensation, blocks where motion compensation is performed and blocks where motion compensation is not performed are mixed. . Parallel movement of the entire screen with a size greater than a certain value (panning; Pannin
In the case where g) or the like occurs, if there is a block that is not motion-compensated surrounded by motion-compensated blocks in a flat portion on the screen, it causes image quality degradation.

In view of the above problems, the present invention checks whether a motion vector of a block that is not motion-compensated by the conventional method is the same as a motion vector of a neighboring motion-compensated block, and determines whether a coded block is panned. In the case of the block at the center of, the determination that no motion compensation is performed is corrected, motion compensation is performed, and the image quality is improved.

Means for Solving the Problems The technical solution of the present invention for achieving the above object is to provide means for converting a television signal from analog to digital, and one frame or one field of a digitized input television signal. Means for dividing the block into blocks of size, means for calculating a motion vector which is the motion of the television image for each block, and whether or not to perform motion compensation on the encoded block based on the inter-frame difference value. First motion compensation determining means for determining the result and calculating the result as a motion compensation control signal; and means for accumulating a motion vector of a coded block and a neighboring block adjacent to the coded block, and a motion compensation control signal. The motion vector and the coding of the coded block determined as not performing the motion compensation by the first motion compensation determining means. A motion compensation control signal correcting means for comparing motion vectors of neighboring blocks adjacent to the block and changing the motion compensation control signal to perform motion compensation when the number of blocks which can be regarded as coincident with the motion vector is K or more; Means for calculating a difference between a pixel value of a coding block and a predicted pixel value based on the corrected motion compensation control signal as a prediction error value, and orthogonally transforming the prediction error value to calculate an orthogonal transformation coefficient Means for quantizing the orthogonal transform coefficients and calculating the quantized orthogonal transform coefficients; means for encoding the quantized orthogonal transform coefficients; and performing inverse orthogonal transform on the quantized orthogonal transform coefficients to perform quantization. Means for calculating a predicted error, a means for calculating a reproduced image from the quantized prediction error and the predicted signal, means for accumulating the reproduced image, and means for encoding a motion vector. The inter-compensated interframe coding device is intended to achieve the above object.

In a conventional motion-compensated interframe coding apparatus, an input television signal is compared with a reproduced television signal of a previous frame stored in an image memory, a motion compensation vector of a coding block is calculated, and a motion compensation vector of the coding block is calculated. The effectiveness of motion compensation is determined using the difference between the previous frame and the current frame, and the difference between the predicted pixel value using motion compensation and the pixel value of the current frame.

However, due to these values, in a portion where a large number of blocks are determined near the boundary of whether or not to perform motion compensation, blocks to be motion-compensated and blocks not to be motion-compensated are mixed. In particular, when there is a block, which is surrounded by motion-compensated blocks and is not motion-compensated, in a flat portion of the screen, such as when the entire screen is translated, this causes deterioration in image quality.

Therefore, according to the present invention, when the motion vector of a coded block is the same as the motion vector of a neighboring block according to the above configuration, even in a block determined not to perform motion compensation, the motion is compensated by correcting the determination. Thus, the image quality is improved.

Embodiment 1 Hereinafter, a first embodiment of the present invention will be described with reference to FIG. FIG. 1 is a block diagram of a motion compensation interframe coding apparatus according to a first embodiment of the present invention.
In FIG. 1, reference numeral 1 denotes an input terminal to which an input television signal is input, and 3 denotes a comparison between an image signal of an encoded block of a current frame and a reproduced image signal of a previous frame, and a motion vector of the encoded block and a first motion. A first motion compensation determination unit for calculating a compensation control signal, an image memory unit for storing a reproduced image signal of a current frame and a previous frame, and a storage unit for storing a first motion compensation control signal and a motion vector. It is determined whether or not motion compensation is to be performed on the converted block, and the result is referred to as a second motion compensation control signal 12.
A motion compensation unit that performs motion compensation on the reproduced image signal of the previous frame, a filter unit 15 that performs two-dimensional low-pass filtering on the second motion compensation control signal, Reference numeral 18 denotes a prediction error calculation unit that calculates a prediction error by performing a difference operation between an original image signal and a prediction signal of a coding block, 20 denotes an orthogonal transformation unit that orthogonally transforms a prediction error signal, and 22 denotes quantization that quantizes orthogonal transformation coefficients. Unit, 25 is an inverse orthogonal transform unit that performs inverse orthogonal transform on the quantized orthogonal transform coefficients, 27
Is a reproduced image calculation unit for calculating the reproduced image of the current frame, 29
Is a prediction error coding unit that performs channel coding on prediction errors, 31 is a motion vector coding unit that performs channel coding on motion vectors, 33
Is a multiplexer unit that forms a transmission frame from a prediction code and a motion vector code, 35 is a code memory unit that temporarily stores the transmission frame, and 37 is an output terminal that outputs a transmission signal.

The operation of the above configuration will be described below. The television signal is analog-to-digital converted by a signal processing unit (not shown in FIG. 1), divided into blocks of M pixels in the horizontal direction and N lines in the vertical direction, and input as a television signal 2 from the input terminal 1. Next, the first
The motion compensation determination unit 3 compares the input television signal 2 with the reproduced image 5 of the previous frame read from the image memory unit 4 and calculates a motion vector 6. At the same time, the first motion compensation determination unit 3 determines whether the motion compensation for the processing block is valid or invalid using the evaluation value at the time of calculating the motion vector, and outputs the result as the first motion compensation control signal 7. The second motion compensation determination unit 8 determines whether the motion vector 6 of the
The motion compensation control signal 7 is accumulated, it is finally determined whether or not the motion compensation is effective for the encoded block, and the result is output to the second motion compensation control signal 12. Further, when it is determined that the motion compensation for the coded block is finally effective, the second motion compensation determination unit 8 outputs a motion vector for the accumulated coded block to the second motion compensation control signal 12.

Therefore, the motion vector and the motion compensation signal are superimposed on the second motion compensation control signal 12.

The final method of determining whether the motion compensation is valid or invalid will be described below.

(1) If the first motion compensation control signal 7 for the coded block is 7ON, the second motion compensation control signal 12 is turned ON and the determination ends.

(2) The motion vector (Vx (i), Vy
(J)) and the motion vectors (Vx (i-1), Vy (j-1)) and (Vx
(I), Vy (j-1)), (Vx (i + 1), Vy (j-
1)), (Vx (i-1), Vy (j)), (Vx (i + 1),
Vy (j)), (Vx (i-1), Vy (j + 1)), (Vx
(I), Vy (j + 1)), (Vx (i + 1), Vy (j +
In 1)), the motion vectors of the reference blocks to be motion-compensated are compared, and the number of blocks in which the motion vectors match is counted. If the number of matches is greater than or equal to a predetermined value K, the first motion compensation is performed. The control signal 7 is compensated, the second motion compensation control signal 12 is turned on, and in other cases, the second motion compensation control signal is turned off.

According to the experiment, improvement of the image quality was confirmed when K = 4.

The motion compensator 14 performs motion compensation on the reproduced image 15 of the coded block using the motion vector when motion compensation is performed using the motion vector information 12, and outputs nothing as the motion compensation signal 16 without performing motion compensation. The in-loop filter unit 15
The in-loop filter processing is performed on the motion compensation signal 16 and the result is output as a prediction signal 17.

Since this processing method requires the motion vectors of eight peripheral blocks adjacent to the in-loop filter processing of the coding block, the processing start of the coding block is delayed by N lines as compared with the conventional example. Processing frame speed; 10F
When RAME / SEC, total number of lines per frame; 288 lines, N = 8, the effect is about 3 mSEC, and the effect is almost negligible.

The prediction error calculation unit 18 performs a difference operation between the input television signal 2 of the coding block and the prediction code 17, and outputs the result as a prediction error signal 19. The orthogonal transform unit 20 performs an orthogonal transform on the prediction error signal 19, and outputs the prediction error signal
The prediction error orthogonal transform coefficient 21 is calculated by removing the high correlation between the 19 neighboring pixels. In many cases, as the orthogonal transform method, a discrete cosine transform that has high conversion efficiency and is feasible in terms of hardware is used. The quantization unit 22 is a quantizer whose quantization characteristic changes according to the generated code amount 23, and quantizes the prediction error orthogonal transform coefficient 21 to calculate a prediction error orthogonal transform quantization coefficient 24. Inverse orthogonal transform unit 25
Calculates the prediction error signal 26 including the quantization error by inversely transforming the prediction error orthogonal transform quantization coefficient 24. The reproduction image calculation unit 27 adds the prediction error signal 26 including the quantization error and the prediction signal 17, and calculates a reproduction image 28 of the encoded block. The image memory 4 stores the reproduced image signal 28 of the current frame and outputs the reproduced image signals 5 and 15 of the previous frame. The prediction error encoding unit 29 is a prediction error orthogonal exchange quantization coefficient.
24 is encoded, and a prediction error code 30 is calculated. The motion vector encoding unit 31 encodes the motion vector 12 of the motion-compensated block, and calculates a motion vector code 32. The multiplexer unit 33 calculates a transmission frame 34 of a predetermined format from the prediction error code 30 and the motion vector code 32. The code memory unit 35 temporarily stores the transmission frame 34 and outputs the transmission frame 34 from the output terminal 37 as a transmission code 36 in synchronization with a clock signal input from the outside (not shown).

As is apparent from the above description, according to the present embodiment, the second motion compensation determination unit 8 compares the motion vector of the coded block with the motion vector of the neighboring block, and determines that the motion vector of the coded block is adjacent. When the motion vector is the same as the motion vector of the surrounding block, the motion compensation is performed also on the block determined not to be motion-compensated. The image quality is improved by preventing the deterioration of the image quality caused by the mixture of the blocks that are not motion compensated.

In the above description, for a block that is determined not to perform motion compensation, whether or not to perform the motion compensation by correcting the determination is determined as “the motion vector of the coded block and the neighboring motion-compensated block adjacent to the coded block. Although the example in which the determination is made based on the “number of matching motion vectors” is shown, the determination may be made based on the “number that can be regarded as matched”.

Further, in the above description, an example in which a motion vector is detected using the reproduced image of the previous frame and the current image of the current frame has been described. However, the motion vector is detected using the original image of the previous frame and the original image of the current frame. , A motion vector may be detected.

As described above, as an effect of the present invention, regarding the determination of the effectiveness of motion compensation, for a coding block determined not to perform motion compensation, the motion vector of the neighboring block adjacent to the motion vector of the coding block is calculated. The temporary storage means is compared with the motion vector of the coded block and the motion vector of the neighboring block. If the motion vector of the coded block is the same as the motion vector of the neighboring block, it is determined that no motion compensation is performed. To prevent the deterioration of image quality caused by the presence of blocks that are motion-compensated and blocks that are not motion-compensated in a flat part of the screen, such as when the entire screen is moved in parallel. As a result, the image quality can be improved, and the effect is great.

[Brief description of the drawings]

FIG. 1 is a block diagram of a motion-compensated interframe coding apparatus according to an embodiment of the present invention, FIG. 2 is a diagram showing a positional relationship between a coding block and neighboring blocks adjacent thereto, and FIG. It is a block connection diagram of a compensation interframe coding apparatus. 1, 41 ... input terminal, 3 ... first motion compensation determination unit, 43 ...
.., A motion vector calculation unit, 8, a second motion compensation determination unit, 4,
44 …… Image memory unit, 14, 48 …… Motion compensation unit, 15, 50…
… Intra-loop filter section, 18, 52 …… Prediction error calculation section, 2
0, 54: orthogonal transform unit, 22, 56: quantization unit, 25, 59 ...
… Inverse orthogonal transform unit, 27, 61… Reproduction image calculation unit, 29, 63…
... Prediction error coding unit, 31, 65 ... Motion vector coding unit,
33, 67 ... multiplexer section, 35, 69 ... code memory section, 37, 71 ... output terminals.

 ────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Ikuo Inoue 1006 Kazuma Kadoma, Kadoma, Osaka Prefecture Matsushita Electric Industrial Co., Ltd. (56) References JP-A 1-2229580 (JP, A) 200789 (JP, A)

Claims (1)

(57) [Claims] A means for converting a television signal from analog to digital; a means for dividing one frame or one field of a digitized input television signal into blocks of a predetermined size; Means for calculating a motion vector which is the motion of the first motion, and determining whether or not to perform motion compensation on the coded block based on the inter-frame difference value, and calculating the result as a motion compensation control signal. Compensation judging means, means for accumulating the motion vector of a coded block and a neighboring block adjacent to the coded block and a motion compensation control signal, and judging that motion compensation is not performed by the first motion compensation judging means. The motion vector of the coded block is compared with the motion vectors of neighboring blocks adjacent to the coded block, and the motion vector A motion compensation control signal correcting means for changing the motion compensation control signal to perform motion compensation when the number of blocks which can be regarded as coincident with each other is equal to or more than K (where K is a positive integer of 8 or less); Means for calculating the difference between the pixel value of the converted block and the predicted pixel value based on the corrected motion compensation control signal as a prediction error value,
A means for orthogonally transforming the prediction error value and calculating an orthogonal transform coefficient; a means for quantizing the orthogonal transform coefficient to calculate a quantized orthogonal transform coefficient; a means for encoding the quantized orthogonal transform coefficient; Means for inversely orthogonally transforming the quantized orthogonal transform coefficients to calculate a quantized prediction error, means for calculating a reproduced image from the quantized prediction error and the prediction signal, means for accumulating the reproduced image, A motion-compensated inter-frame encoding device comprising encoding means.