JPH0410884A - Moving picture encoding transmitting device - Google Patents

Abstract

PURPOSE:To reproduce a more natural moving picture with smooth movement and small, distortion by removing a high frequency distortion component with lowering cut-off frequency when the encoding distortion is large and reproducing an original high frequency signal component with returning the cut-off frequency to a normal value when the encoding distortion is small. CONSTITUTION:An encoder 24 compress-encodes a differential signal between flames from a subtracter 8 according to fixed algorithm and a distortion value signal generating part 25 is built in the encoder 24 and generates an encoding distortion value signal to indicate the quantity of encoding distortion generated at encoding. Besides, a receiver 2 provides a low-pass filter 28 able to control cut-off frequency to filter a decoding video signal and controls the cut--off frequency of the low-pass filter 28 based on the encoding distortion value signal to indicate the generating quantity of the encoding distortion generated at the encoding. That is, when the encoding distortion is large, a high frequency distortion component is removed with lowering the cut-off frequency and when the encoding distortion is small, an original high frequency signal component is reproduced with returning the cut-off frequency to a normal value. Thus, a more natural moving picture can be obtained.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a moving picture encoding and transmitting apparatus that compresses and encodes a video signal of a moving picture and transmits the compressed and encoded video signal.

[Conventional technology]

Figure 8 shows, for example, a conventional moving image shown in the paper "Teleconferencing System" published on pages 38 to 41 of Mitsubishi Electric Branch Volume 60, No. 10 (published by Mitsubishi Electric Corporation in 1986). FIG. 2 is a block diagram showing an encoding and transmitting device. In the figure, 1 is a transmitting device that encodes and transmits a video signal of an input moving image for each encoded video frame, 2 is a receiving device that receives and decodes the encoded video signal, and 3 is a transmission line connecting the transmitting device 1 and the receiving device 2.

4 is a video signal input terminal pl for inputting an input video signal to the transmitting device 1; 5 is a video signal input terminal υ
This is an analog input section that digitizes the input video signal and performs luminance/chrominance component separation processing. Reference numeral 6 denotes a transmission frame memory that stores video data for one frame processed by the analog input section 5, and reference numeral 7 designates a frame memory in which video data of the previous frame is stored. Reference numeral 8 denotes a subtracter that calculates the difference between the video data of the current frame read from the transmission frame memory 6 and the video data of the previous frame read from the frame memory 7.

9 is an encoder that compresses and encodes the interframe difference signal from the subtracter 8 according to a predetermined algorithm;
10 is a transmission buffer that temporarily stores encoded data from the encoder 9. Reference numeral 11 designates a transmission interface section that transmits the encoded data stored in the transmission buffer 10 in accordance with the transmission speed of the transmission line 3; This is an output terminal for encoded data to be output.

Decoders 1 and 3 decode the encoded data from the encoder 9 in the reverse procedure of the encoder 9 to reproduce an interframe difference signal. 14 adds the video data of the previous frame stored in the frame memory 7 to the inter-frame difference signal reproduced by the decoder 13 to generate the video data of the current frame, which is used for the next frame. This is an adder that re-stores the inter-frame difference signal in the frame memory 7 for calculation.

15 is a receiving device 2 for receiving encoded data from the transmission line 3.
16 is an input terminal for encoded data to be input to the transmission line 3, and a reception interface section 16 receives the encoded data in accordance with the transmission speed of the transmission line 3.

17 is a reception buffer that temporarily stores the encoded data received by the reception interface section 16, and 18 is a reception buffer that temporarily stores the encoded data stored in the reception buffer 1T.
This is a decoder that performs decoding using the reverse procedure of the encoder 9 of , and restores the interframe difference signal.

19 is a frame memory in which the video data of the previous frame is stored, and 20 is the frame memory in which the video data of the previous frame stored in this frame memory 19 is added to the interframe difference signal restored by the decoder 18. This is an adder that performs addition to restore the video data of the current frame. This adder 2
The video data restored at 0 is stored again in the frame memory 19 for restoring the data of the next frame.

Reference numeral 21 denotes a receiving frame memory that stores the video data of the current frame restored by the adder 20, and 22 sequentially reads out the video data stored in the receiving frame memory 21 at a speed according to the video rate of the video signal. This is an analog output section that converts signals into analogs. 23 is a video signal output terminal to which the video signal from this analog output section 22 is output.

Next, the operation will be explained. Video signal input terminal 4
A video signal input to the transmitter 1 is converted from an analog signal to a digital signal by an analog input section 5, separated into a luminance component and a color difference component, and then written into a transmit frame memory 6. In the transmission frame memory 6, the video rate of the input video signal (for example, NTSC
30 frames per second in the color television system) and the encoded video rate in the subsequent encoder 9,
Scan conversion is performed from the rask scan to a format suitable for encoding.

After that, the video data of the current frame read from the transmission frame memory 6 and the video data of the previous frame read from the frame memory 7 are sent to the subtracter 8, and the difference between them is calculated, and the inter-frame difference is calculated. Output as a signal. Since this inter-frame difference signal is a change component between the video data of the previous frame and the video data of the current frame,
Its absolute value is much smaller than the video data itself of the current frame, and the amount of υ information is reduced.

This interframe difference signal is input to the encoder 9 and compressed and encoded based on a predetermined encoding algorithm, thereby further reducing the amount of information. This encoding result is temporarily stored in the transmission buffer 10 as encoded data, and then
According to the transmission rate determined by the transmission line 3, the encoded data is sent to the transmission line 3 via the transmission interface section 11 and the encoded data output terminal 12.

On the other hand, the encoded data encoded by the encoder 9 is subjected to decoding processing in a reverse procedure to the encoding in the decoder 13 in preparation for the encoding operation of the next frame, and an interframe difference signal is generated. is played as. This reproduced inter-frame difference signal is added to the video data of the previous frame outputted from the frame memory 7 by the adder 14, and is stored again in the frame memory T as the video data of the current frame. When encoding the next frame, this video data is used as the previous frame's video data for encoding according to the procedure described above.

The transmitting device 1 operates as described above.

Next, the encoded data transmitted via the transmission line 3 is input to the receiver 2 through the encoded data input terminal 15,
υ to the transmission line 3 via the reception interface section 16
The data is written to the reception buffer 17 according to the determined transmission rate. Thereafter, the received encoded data undergoes decoding processing in a decoder 18 having the same configuration as the decoder 13 of the transmitting device 1, and is restored as an interframe difference signal. This interframe difference signal after restoration is added to the video data of the previous frame outputted from the frame memory 19 by the adder 20, and is restored as the video data of the current frame.

The restored video data is once written into the reception frame memory 21, read out sequentially at a speed of -E+7 according to the picture rate of the video signal, and sent to the analog output section 22.

The analog output section converts it into an analog signal, reproduces it as a video signal, and outputs it from the video signal output terminal 23.

On the other hand, the video data of the current frame restored by the adder 20 is written into the frame memory 19 again, and is used as the video data of the previous frame when decoding the next frame.

The receiving device 2 operates as described above.

Next, the operation of encoder 9 will be described. The inter-frame difference signal input to the encoder 9 has a much smaller value υ than the video data itself of the current frame mentioned above, but if all information is to be transmitted, the amount of generated information will be far exceeds the capacity of the transmission line 3.

For example, if a vector quantization method is used as the encoding method,
Consider a case where one screen is composed of 336 pixels x 240 lines. Here, for example, spectrum Vo1.01
As shown in Figure 13 of , NO. 05, 1988-05P99, conversion patterns are obtained from the codebook with one block having a configuration of 4 pixels x 4 lines, and the number of patterns stored in the codebook in advance is 4096. If it is a pattern, 4096=2'', so 12 bits are required to express one block.

This block group consisting of 1 block of 4 pixels x 4 lines does not exist in the 336 pixels x 240 lines, and if 1 block is expressed with 12 bits, it becomes 5040 x 12 = 60,480 bits of data. will occur.

For example, convert this 60,480 bit data into 64K
When transmitted over the bps transmission line 3, the number of frames transmitted per second is as follows, and it can no longer be called a moving image.

For this reason, the encoder 9 preferentially encodes areas with more rapid changes within one screen (areas with large inter-frame difference signals), and uses the value of the video data of the previous frame for areas with less changes. !!, by making a decision to use i and encoding and transmitting only, for example, 500 blocks out of the 5,000+ blocks mentioned above, the number of transmitted frames per second can be increased, resulting in a more natural video. However, since the input video signal changes from moment to moment, the magnitude of the interframe difference signal input to the encoder 9 also decreases by 1. It changes every frame.In order to select a certain number of blocks from among these, the threshold for determining whether to encode or not is also changed every frame.When the change is rapid, the threshold is raised and the When the number of frames is low, the threshold is controlled to be lowered to suppress fluctuations in the number of frames transmitted per second.

[Problem to be solved by the invention]

Since the conventional video encoding and transmission device is configured as described above, it is possible to secure a constant number of transmission frames even when the input video signal changes, and to transmit and reproduce video with more natural motion. However, as the threshold for determining whether or not to perform encoding changes, the image quality of the reproduced image also changes for each video frame. When this happens, areas where a certain amount of change has occurred, although not enough to reach the determination threshold, are not encoded, resulting in an increase in final encoding distortion. If using
Since coding distortion tends to occur in high-frequency components that are easily visible, the determination threshold cannot be increased unnecessarily, and the final number of frames to be transmitted must be limited to a certain value or less.

This invention was made in order to solve the above-mentioned problems, and because the change in the input image is large, the threshold value for determining whether or not to perform encoding shows a high value, and the encoded image data Even if high-frequency encoding distortion increases, the high-frequency distortion component in the video signal that is finally output from the device is removed, resulting in a more natural moving image with little change in the input video signal. Sometimes, the objective is to obtain a moving image encoding and transmitting device that operates to output the original high-frequency signal in an input video signal.

[Means to solve the problem]

The video encoding and transmitting device according to the present invention includes a transmitting device that includes a distortion value signal generation unit that generates an encoded distortion value signal indicating the amount of encoding distortion that occurs when encoding a video signal, and and a multiplexing unit that multiplexes the encoded video signal into the encoded video signal, and the receiving device includes a separating unit that separates the multiplexed encoded distortion value signal and the encoded video signal, and a separated encoding unit. It is equipped with a low-pass filter for filtering the decoded video signal, the cut-off frequency of which is controlled by the distortion value signal.

[Effect]

The receiving device according to the present invention is equipped with a low-pass filter whose cut-off frequency can be controlled for filtering a decoded video signal, and includes a code indicating the amount of encoding distortion generated during encoding, which is sent from the transmitting device. The cutoff frequency of this low-pass filter is controlled based on the encoded distortion value signal, and when the encoding distortion is large, the cutoff frequency is lowered to remove high-frequency distortion components, and when the encoding distortion is small, the cutoff frequency is lowered to the normal frequency. By returning to the value of pl and reproducing the original high frequency signal component,
A moving image encoding and transmitting device capable of obtaining very natural moving images is realized.

〔Example〕

An embodiment of the present invention will be described below with reference to the drawings. 1st
In the figure, 1 is a transmitting device, 2 is a receiving device, 3 is a transmission line, 4 is a video signal input terminal, 5 is an analog input section, 6
is a transmission frame memory, T is a frame memory, 8 is a subtracter, 11 is a transmission interface section, 12 is an encoded data output terminal, 13 is a decoder, 14 is an adder, 15 is an encoded data input terminal, 16 is a Reception interface section, 18
is a decoder, 19 is a frame memory, 20 is an adder, 2
1 is a reception frame memory, 22 is an analog output section, 23
1 is a video signal output terminal, and since it is the same as or corresponds to the conventional terminal with the same reference numerals in FIG. 8, detailed explanation thereof will be omitted.

24 is an encoder that compresses and encodes the interframe difference signal from the subtracter 8 according to a predetermined algorithm; 25;
is a distortion value signal generation section that is built in the encoder 24 and generates an encoded distortion value signal indicating the amount of encoding distortion generated during encoding. This encoder differs from the conventional encoder 9 shown in FIG. 8 in that it has a built-in section 25. Reference numeral 26 denotes a transmission buffer as a multiplexing unit that multiplexes the encoded data from the encoder 24 with the encoded distortion value signal generated by the distortion value signal generation unit 25 and temporarily stores the multiplexed signal.

2T is a receiving buffer serving as a separating section that temporarily stores the multiplexed encoded data and the encoded distortion value signal received by the receiving interface section 16, and separates and outputs them. 28 is a low frequency drive whose cutoff frequency is controlled according to the encoded distortion value signal separated by the reception buffer 27, and which filters the reconstructed video data based on the interframe difference signal decoded by the decoder 18. Pass filter.

Next, the operation will be explained. Here, FIG. 2 is a characteristic diagram showing the frequency characteristics of this low-pass filter 28, FIG. 3 is an explanatory diagram showing frequency components of a video signal when encoding distortion occurs, and FIG. FIG. 3 is an explanatory diagram showing the frequency components of a signal obtained by passing the video signal having the frequency components shown in the figure through a low-pass filter 28;

In the transmitting device 1, the input video signal is digitized in the analog input section 5 and then compressed and encoded in each subsequent stage of processing, and in the receiving device 2, it is digitized from the transmission line 3 through the receiving interface section 16. The operations from sequentially restoring the received encoded data to outputting it as video data from the adder 20 are the same as in the conventional case.

Here, in the encoding process, the encoder 24 performs an operation to determine whether to preferentially encode a portion with a large change among the pixels to be encoded in one screen, as in the conventional case. The distortion value signal generation unit 25 outputs information on the threshold value for determining whether or not to perform encoding in the encoder 24 as an encoded distortion value signal.

This encoded distortion value signal is sent to the transmission buffer s26,
It is multiplexed and temporarily stored together with the encoded data from the encoder 24, and is outputted to the transmission line 3 from the encoded data output terminal 12 via the transmission interface section 11.

On the other hand, in the receiving device 2, the multiplexed encoded data and the encoded distortion value signal received by the reception interface section 16 are temporarily stored in the reception buffer 27, and are separated to decode the encoded data. The encoder 18 sends the encoded distortion value signal to the low-pass filter 28 .

The encoded data input to the decoder 18 is restored to an interframe difference signal in the same manner as in the conventional case, as described above. Furthermore, the encoded distortion value signal input to the low-pass filter 28 acts as control information for the low-pass filter 28 to change its cutoff frequency.

FIG. 2 shows how the cut-off frequency of the low-pass filter 28 is changed. Normally, as shown by the solid line, the low-pass filter 28 has a low-pass characteristic with the cut-off frequency being f0. (H2) Attenuate higher frequency components. Furthermore, when the input encoded distortion value signal shows a large value, the cutoff frequency is changed to f+cfx<fo), and f
l (H2: Attenuate frequency components of 1 or higher.

Next, paying attention to the nature of the restored video data input to the low-pass filter 28, the low-pass filter 28
Explain the operation. FIG. 3 shows the distribution of frequency components of reconstructed video data that is multiple inputted to the low-pass filter 28 by the adder 20.

In the transmitting device 1, when there are many changes between frames of the input video signal, the threshold for determining whether or not to perform encoding is increased in order to suppress an increase in the amount of encoded information, as described above, as in the conventional case. As a result, the area in which there is actually a change but is not encoded increases in the screen, and as a result, a large amount of error from the original image occurs as encoding distortion. Regarding the frequency component of the encoding distortion at this time, if vector quantization similar to the conventional case is used as the encoding method, noticeable distortion tends to occur in the high frequency portion of the video signal. FIG. 3 shows that this high-frequency distortion component exists in the range f1 to fo.

When the signal having the spectrum shown in FIG. 3 is passed through the low-pass filter 28 having the characteristics shown in FIG. As a result, as shown in Fig. 4, a signal with attenuated high-frequency components of encoding distortion is obtained.The signal from which this high-frequency distortion component has been removed is stored in the reception frame memory 21 and output as an analog output. The signal is analog-converted in the unit 22 and outputted from the video signal output terminal 23. Therefore, it is possible to obtain a reproduced moving image with improved image quality.

In the above embodiment, the cutoff frequency of the low-pass filter 28 is f. In the above example, the value of the encoded distortion value signal changes continuously, and various cutoff frequencies are changed according to the value of the encoded distortion value signal. Alternatively, the image quality may be changed continuously, and in this case, there is an effect that sudden changes in image quality for each frame can be prevented.

Furthermore, in the above embodiment, a threshold value for determining whether or not to perform encoding is used as the encoding distortion value signal in the encoder 24, but the decoder 13 of the transmitter 1 It is also possible to find the error between the interframe difference signal decoded and the interframe difference signal input to the encoder ν4, and use the cumulative value of 9 for each video frame of the error signal as the encoded distortion value signal. good.

FIG. 5 is a block diagram showing an example of the configuration of the transmitter ff1 in such an embodiment. In the figure, 29 is an interframe difference signal locally decoded by the decoder 13;
The subtracter 30 detects an error in the inter-frame difference signal input to the encoder 9, and the accumulator 30 accumulates the error detected by the subtracter 29 for one video frame. The distortion value signal generation section 25 is formed by the subtracter 29 and the accumulator 30.

In the transmitting device 1 configured in this way, the decoder 1
Since the inter-frame difference signal locally decoded in step 3 contains coding distortion, the difference between the inter-frame difference signal input to the subtracter 29 and the inter-frame difference signal input to the encoder 9 is calculated as follows. The subtracter 29 outputs only the encoded distortion component. By integrating the coding distortion components for one video frame in the accumulator 30, it is possible to directly determine the extent to which coding distortion occurs in the video frame.

The distortion value signal generation unit 25 notifies the transmission buffer 26 of the encoded distortion value signal. The subsequent operations in the transmitter 1 and the receiver 2 are the same as in the embodiment shown in FIG.

Furthermore, in the embodiment shown in FIG. 1, the low-pass filter 28 whose cut-off frequency can be controlled is arranged before the reception frame memory 21, but there is a low-pass filter 28 placed between the reception frame memory 21 and the analog output section 22. Alternatively, it may be placed inside the analog output section 22.

FIG. 6 is a block diagram showing an example of the configuration of the analog output section 22 in an embodiment in which the low-pass filter 28 is arranged within the analog output section 22. In the figure, 3
1 is a DA converter that converts received video data into analog, 32 is a plurality of individual low-pass filters arranged in parallel, each with a different cutoff frequency, and 33 is a coded distortion filter. This is a switch that selects one of the outputs of each individual filter 32 based on the signal. The low-pass filter 28 whose cut-off frequency can be changed is composed of the plurality of individual filters 32 and switches 33.

In the analog output section 22 configured in this way, when the video data from the reception frame memory 21 is input to the DA converter 31, it is converted into an analog signal and sent to each individual filter 32 of the low-pass filter 28. input in parallel. Each individual filter 32h performs filtering at a different cutoff frequency, and removes frequency components above the cutoff frequency. On the other hand, the switch 33 selects the output from the individual filter 32 with a low cut-off frequency when the encoded distortion value signal shows a large value, and selects the output from the individual filter 32 with a high cut-off frequency when the encoded distortion value signal shows a small value. Its behavior is configured to select the output. Therefore, it operates in the same manner as the embodiment shown in FIG. 1, and reproduced moving images with improved image quality can be obtained.

Further, in FIG. 6, a plurality of individual filters 32 are arranged in parallel and one of them is selected by the switch 33, but as shown in FIG. 7, nine individual filters 32 are arranged in series.
The switch 33 may be used to select the output.

〔Effect of the invention〕

As described above, according to the present invention, the transmitter transmits an encoding distortion value signal indicating the amount of encoding distortion generated during encoding, and the receiver uses a cutoff frequency to filter the decoded video signal. The cutoff frequency of the low-pass filter, which can be controlled, is controlled by this encoded distortion value signal,
When the coding distortion is large, the cut-off frequency is lowered to remove high-frequency distortion components, and when the coding distortion is small, the cut-off frequency is returned to the normal value to reproduce the original high-frequency signal components. Even if the high-frequency distortion component in the encoded data increases due to many changes in the input video signal, it will not appear in the playback output, and this increases the threshold for determining whether to perform υ encoding. Since it becomes possible to increase the number of transmitted frames, it is possible to obtain a video encoding and transmitting device that can reproduce natural video with smooth motion and less distortion. effective.

[BRIEF DESCRIPTION OF THE DRAWINGS] FIG. 1 is a block diagram showing a moving picture encoding/transmission apparatus according to an embodiment of the present invention, FIG. 2 is a characteristic section showing the frequency characteristics of the low-pass filter, and FIG. FIG. 4 is an explanatory diagram showing the frequency components of a video signal when encoding distortion occurs; FIG. 4 is an explanatory diagram showing the frequency components of a signal obtained by passing the video signal with the frequency components shown in FIG. 3 through a low-pass filter; FIG. 5 is a block diagram showing the configuration of a transmitter according to another embodiment of the invention, FIGS. 6 and 7 are block diagrams showing the configuration of an analog output section according to still another embodiment of the invention,
FIG. 8 is a block diagram showing a conventional video encoding and transmitting device. 1 is a transmitting device, 2 is a receiving device, 3 is a transmission line, 9.24
is an encoder, 18 is a decoder, 25 is a distortion value signal generation unit,
26 is a multiplexing unit (transmission buffer), 27 is a demultiplexing unit (reception buffer), and 28 is a low-pass filter. In addition, the same reference numerals in the figures indicate the same parts or priest parts. Patent applicant: Mitsubishi Electric Corporation (2 others) G +ds+ Ushizo Ryousuke 1st grade Figure 7 Narov = Tokaku Yang x

Claims (1)

[Claims] The video signal of the input moving image is converted into 1 by the encoder.
A moving image comprising: a transmitter that encodes each encoded video frame and sends it to a transmission line; and a receiver that decodes the encoded video signal received from the transmission line using a decoder and outputs the decoded video signal. In the encoding and transmitting device, a distortion value signal generation unit that generates an encoding distortion value signal indicating the amount of encoding distortion generated when the input video signal is encoded by the encoder in the transmitting device. and a multiplexing unit that multiplexes the encoded distortion value signal generated by the distortion value signal generation unit onto the video signal encoded by the encoder, and the receiving device includes a a separation unit that separates the coded distortion value signal and the coded video signal multiplexed in the separation unit; and a cutoff frequency thereof is controlled in accordance with the coded distortion value signal separated in the separation unit. and a low-pass filter that filters the video signal decoded by the decoder.