JP2897921B2 - Image coding transmission equipment - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Summary] The present invention relates to an image encoding and transmitting apparatus that divides a screen into a plurality of blocks, encodes the blocks, and transmits the encoded image signals through a plurality of channels having different discard rates. Even if several blocks of image signals are lost due to cell discard in the transmission path, the resulting deterioration of the playback screen is limited to only the lost blocks, and the image degradation is widespread. An image encoding and transmitting apparatus for transmitting an image signal encoded in units of blocks through a plurality of channels provided on a network side and having different discard rates. It is configured to transmit the image information of the block to the side channel and to transmit the block configuration information of the image information of the channel to the high loss rate channel to the channel with the low loss rate. It is.

[Industrial applications]

The present invention relates to an image coding transmission apparatus that divides a screen into a plurality of blocks, codes the blocks, and transmits the coded image signals through a plurality of channels having different discard rates.

In such an image encoding / transmission apparatus, it is necessary to prevent the reproduction screen from being distorted as much as possible on the receiving side even when a part of the transmission image information is discarded in the network.

[Conventional technology]

In a high-efficiency encoding method for image signals, when an encoded image signal is transmitted using a burst multiplexing transmission network such as ATM or packet switching, the transmission frame is lost due to a transmission error in a transmission path, a buffer overflow, or the like. (Discarding of cells or packets) is inevitable.

As a method to cope with this, a plurality of channels having different discard rates are provided on the network side, and these channels are prioritized. The image information having high importance is transmitted through a channel having a low discard rate, and the image information having low importance is transmitted through a channel having a high discard rate. A multi-channel coded transmission scheme has been proposed in which the discard rate is such that information is discarded first from the channel side and the minimum reproduction picture quality on the receiving side can be guaranteed even for cell discard on the transmission line. ing.

[Problems to be solved by the invention]

As shown in FIG. 4, in an image coding apparatus that performs high-efficiency coding of an image signal in block units, an image signal that has been A / D-converted by a pre-processing unit (A / D conversion unit) 21 is converted into a source code. The encoding unit 22 encodes each block, and the source encoding unit
The coded image signal from 22 is further variable-length coded together with various block attribute information in a video multiplex encoder 23 and multiplexed into a predetermined coded sequence, and the coded image signal sequence is multiplexed with an audio signal or the like. To the transmission path via the multiplexing block 24.

FIG. 5 shows the frame configuration of the image signal coded sequence. As shown in the figure, a unique pattern (unique word UW) for synchronization detection is inserted at the beginning of a processing unit such as a frame or a sub-block, followed by a block line header, and then a block of each block. Attribute information and image information are arranged continuously.

When a cell is discarded with respect to an image signal having such a frame configuration, the positional relationship of the screen of the block information after the lost portion is disturbed. Decoding of subsequent blocks becomes impossible, and basically decoding becomes impossible until a unique word UW of the next processing unit arrives. As a result, assuming that a unique word is provided for each frame, for example, as shown in FIG. 6, the screen quality is deteriorated over a wide range (hatched portion) after the location where the cell was discarded.

Therefore, an object of the present invention is to make it possible that even if an image signal of several blocks is lost due to cell discarding on a transmission line or the like, the resulting deterioration of the reproduced screen is limited to only the lost block portion. An object of the present invention is to prevent image quality deterioration from spreading over a wide range.

[Means for solving the problem]

An image encoding and transmitting apparatus according to the present invention is an image encoding and transmitting apparatus for transmitting an image signal encoded in units of blocks through a plurality of channels provided on a network side and having different discard rates. The image information in the block unit is transmitted to the channel on the discard rate side, and the image information in the coded data string transmitted on the channel on the high discard rate side is transmitted to the channel on the low discard rate at which block position in the reproduction screen. It is configured to transmit configuration information (hereinafter, referred to as block configuration information in the present specification) for determining whether a block is a device.

According to another aspect of the present invention, there is provided an image encoding transmission apparatus according to the above-described image encoding transmission apparatus, which separates pixel information into high-frequency components and low-frequency components and discards low-frequency component pixel information. And transmitting the high frequency component pixel information via the high discard rate channel.

[Action]

On the network side, a plurality of channels (for example, two channels) are prioritized, and a channel having a low information discard rate (upper channel) and a higher channel (lower channel) are provided. Transmission.

FIG. 1 shows an example of the signal format of the transmission image signal in this case.

On the transmitting side, one screen is divided into a number of blocks, encoding is performed in block units, and pixel information P is generated for each block. In the lower channel, only the image information P ₁ , P _2, ... Of each block is sequentially transmitted. On the other hand, in the upper channel, the pixel information P ₁ , P ₂ ,.
Are transmitted (for example, the number of bits of pixel information P ₁ , P _2, ..., Etc.) of the blocks respectively corresponding to.

As a result, even if the block pixel information P of the lower channel is lost for several blocks due to cell discarding on the transmission line, the receiving side determines which position on the screen based on the block configuration information received via the upper channel. It can be known whether the pixel information of the block has disappeared. Therefore, the correct arrangement position on the screen can be known for the blocks subsequent to the information discarded block, and the image information can be reproduced by decoding the pixel information of these blocks. Therefore, the range of the image quality deterioration can be limited to only the information discarded blocks, and the blocks subsequent to the information discarded blocks do not cause the image quality deterioration.

In another embodiment of the present invention, the pixel information P of the block is converted to a more important component (for example, a low-frequency component) when reproducing an image.
P ′ is separated into a less important component (for example, a high-frequency component) P ″, and the more important component P ′ is transmitted via the upper channel. In this way, information about a certain block in the lower channel is obtained. Even if discarded, only the less important pixel information component P ″ is lost, so that the receiving side can receive the pixel information P ′ for this block portion via the upper channel. Based on the pixel information P ', an image can be reproduced even if the sharpness is slightly lost. As a result, it is possible to suppress deterioration of the image quality of the block portion where the information is discarded.

〔Example〕

 Hereinafter, embodiments of the present invention will be described with reference to screens.

An image encoding and transmitting apparatus according to an embodiment of the present invention
Shown in the figure. In this embodiment, the apparatus of the present invention is used in a multi-channel coded transmission system in which channels with high and low discard rates are provided by assigning priorities to channels on the network side. I have.

In Figure 2, 1 is the discrete cosine transformer to coefficients of frequency components by the discrete cosine transform an image input block, 2 and the output of the discrete cosine transformer 1 low frequency component f _L and the high frequency component f _H And 2 are high-frequency components f _H
Intraframe encoder consisting quantizer for encoded frame, 4 is an inter-frame encoder for interframe coding the low frequency component f _L. The configuration of the inter-frame encoder 4 is a known one, and includes a subtractor 41, a quantizer 42, and an inverse quantizer.
43, an adder 44, a frame memory 45 and the like.

Reference numeral 5 denotes a video multiplex encoder for performing variable length encoding on the output of the intra-frame encoder 3, and its output signal is transmitted to the network via a channel (lower channel) having a high discard rate. Reference numeral 6 denotes a video multiplex encoder which receives the output of the inter-frame encoder 4 and the output of the intra-frame encoder 3 and performs variable-length encoding of the encoded image information together with various block attribute information. The output signal is sent to the network via a channel with a low discard rate (upper channel).

The operation of this embodiment will be described below with reference to FIG. A block image signal obtained by dividing one screen into a large number of blocks is input to the discrete cosine transformer 1, where it is subjected to a discrete cosine transform to be converted into a frequency component. This frequency component is further separated by a coefficient separator 2 into a low-frequency component f _L and a high-frequency component.
is separated into the f _H, the low-frequency component f _L is inter-coded by inter-frame encoder 4, the high frequency component f _H is intraframe coding in intra-frame encoder 3, the video multiplex each The signals are input to the encoders 6 and 5 and multiplexed into an encoded sequence.

FIG. 3 is a diagram showing an example of a frame configuration of a multiplexed coded sequence in an upper channel and a lower channel. In this example, the channel side and the lower channel side are both
Conditional pixel replenishment is performed on a block-by-block basis, and invalid information on the lower channel side is Lang-length coded.

Here, the conditional pixel replenishment calculates an error between a block to be coded on the current screen and a block at the same position on the previous screen, and when the error is smaller than a predetermined threshold, the motion of the screen is Only the "invalid" information is sent as almost no information, and the receiving side uses the pixel information of the block at the same position on the previous screen as the reproduction screen in the case of "invalid". If the error is larger than the threshold value, "valid" information is sent, and pixel information of the block is sent following this "valid" information.

First, in the upper channel, the block on the low frequency coefficient side (the output side of the interframe encoder 4) is "valid" or "invalid".
Or send information. In this case, if "valid",
As in the first block, the third block, etc., the low frequency coefficient information,...
Only send "invalid" information, like blocks.

Further, in the upper channel, the block configuration information of the block of the high frequency coefficient (output from the intra-frame encoder 3) on the lower channel side is transmitted. As block configuration information,
The information is “valid” or “invalid”. If “valid”, it is information including the number of bits of the high-frequency coefficient block. However, the invalid information of the lower channel is run-length coded, that is, if the run length is "0", it indicates that the high frequency coefficient information is "valid". In this case, the number of bits of the lower channel is sent together. If the run length is a finite number, it indicates that the high-frequency coefficient information is "invalid" over the run length after that block. The example of FIG. 3 shows a case where the run length is “98” in the second block and the second to 99th blocks are “invalid”.

In the lower channel, only high frequency coefficient information is sent. However, high frequency coefficient information is not sent for a block determined to be “invalid”. In the example of FIG. 3, since the second block to the 99th block are "invalid", no high frequency coefficient information is sent.

If the low-frequency coefficient information, its attribute information, and the block configuration information of the high-frequency coefficient information are transmitted through the upper channel and the high-frequency coefficient information is transmitted through the lower channel as described above, for example, the transmission frame in the network may be transmitted. Even if the high-frequency coefficient information of the lower channel with a high discard rate is lost due to the disappearance of the high-frequency coefficient information, the high-frequency coefficient information is lost due to the bit number information of the lower channel sent by the upper channel and the run length of the invalid block of the lower channel. The reproduction position on the screen after the second block is not disturbed, and
The lower channel information after the block can be decoded.

In order to detect the position of the lost block, disorder in the coding rule of the block can be used. That is, when decoding a block, if the coding rule at the time of encoding is disordered, it can be estimated that the block has been lost from that position. In this case, the end of the block erasure can be determined by following the coded sequence from the position of the unique word in the next frame in the reverse direction according to the lower channel bit number information from the upper channel.

On the receiving side, upon detecting the loss of the block, decoding is performed by inserting zero information into the discarded data portion. In this way, when a cell is discarded, only the high-frequency components of the discarded data portion are decoded in a lost form, and the effect of the cell discard does not affect other blocks. In addition, since the low-frequency coefficient information is transmitted via the upper channel also for the discarded data portion, it is possible to reproduce an image having the minimum image quality even if the image is not clear.

Note that if the network side inserts dummy data into the discarded data portion and adjusts the number of bits so that the data amount is kept constant between the transmitting and receiving terminals, decoding can be performed more easily.

Various modifications are possible in implementing the present invention. For example, in the above-described embodiment, two-channel transmission is used as the number of channels in the multi-channel coding method. However, the present invention is not limited to this, and three or more channels can be used as the number of channels. For example, in the case of three channels, the channels are prioritized in the order of an upper channel, a middle channel, and a lower channel, and the image information is divided into upper, middle, and lower priority information. , The block configuration information of the lower channel and the middle pixel information are transmitted in the middle channel, and the block configuration information and the upper pixel information of the middle channel are transmitted in the upper channel. Further, the multi-channel coded transmission system is not limited to that of the embodiment, and various systems can be used. Further, the high-efficiency coding method of the pixel information is not limited to the inter-frame coding method and the intra-frame coding method of the embodiment. For example, instead of the intra-frame coding method, a leaky integration type inter-frame coding method is used. Modifications such as using a method are also possible.

〔The invention's effect〕

According to the present invention, even when an image signal of several blocks is lost due to loss of a transmission frame (discard of cells, packets, etc.), the resulting deterioration of the reproduced screen is minimized only to the lost block portion. Therefore, the image quality can be prevented from being deteriorated over a wide range.

[Brief description of the drawings]

FIG. 1 is a diagram for explaining the principle according to the present invention, FIG. 2 is a block diagram of an image encoding and transmitting apparatus as one embodiment of the present invention, and FIG. FIG. 4 is a diagram showing a frame configuration example of a lower channel, FIG. 4 is a block diagram showing a configuration example of a conventional image coding transmission apparatus, FIG. 5 is a diagram showing a conventional multiplexed coded sequence, and FIG. It is a figure for explaining the problem of. In the figure, 1 ... a discrete cosine transformer 2 ... a coefficient separator 3 ... an intra-frame encoder 4 ... an inter-frame encoder 5, 6 ... a video multiplex encoder 41 ... a subtractor 42 ... … Quantizer 43 …… dequantizer 44 …… adder 45 …… frame memory

────────────────────────────────────────────────── ─── Continuation of front page (72) Inventor Kiichi Matsuda 1015 Kamiodanaka, Nakahara-ku, Kawasaki City, Kanagawa Prefecture Inside Fujitsu Limited (56) References JP-A-63-73786 (JP, A) (58) Fields investigated (Int.Cl. ⁶ , DB name) H04N 1/415 H04N 7/133 H04L 11/20

Claims (3)

(57) [Claims] 1. An image signal encoded in a block unit,
An image coding / transmission apparatus provided on a network side for transmitting through a plurality of channels having different discard rates, wherein a high discard rate channel transmits image information in block units and a low discard rate channel An image encoding transmission apparatus configured to transmit configuration information for determining which block position in the reproduction screen each image information in the encoded data sequence transmitted on the channel on the high discard rate side . 2. The method according to claim 1, wherein the image information is separated into a high frequency component and a low frequency component, the low frequency component image information is transmitted through a low discard rate channel, and the high frequency component image information is transmitted through a high discard rate channel. The image encoding transmission device according to claim 1, wherein the image encoding transmission device is configured as follows. 3. The configuration information includes information indicating whether image information of each block can be regarded as being the same as image information at the same position on the previous screen, and if not, bit number information of image information of the block. 3. The image encoding and transmitting apparatus according to claim 1, comprising: