JPH0818958A - Encoder/decoder for video signal - Google Patents

Abstract

PURPOSE:To provide an encoding signal capable of interpolating defective data without using a specific means when an error occurs in the case an image signal is transmitted after applying encoding processing. CONSTITUTION:A video signal inputted from a video signal input terminal 1 is separated to even-numbered and odd-numbered field strings by an even/odd numbered field separation part 2. Each field string receives the encoding processing independently by an even-numbered field encoder part 3 or an odd- numbered field encoder part 4, respectively and forms a GOP, and it is sent to video multiplexer 5. The video multiplexer 5 unifies the GOPs in one signal for transmission, and outputs it from a composite signal output terminal 6, and the signal is transmitted via, for example, a digital network, etc. Also, in the case the error occurs in the transmission, it is detected by a decoder, and after decoding processing is applied to it, interpolation processing is performed, then, it can be restored to an original image signal.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a video signal coding apparatus for coding an input video signal, and the present invention relates to a video signal decoding apparatus for decoding a coded video signal.

[0002]

2. Description of the Related Art In recent years, due to the development of cables and the like, a service for handling reduced-size images as a means for communicating images, so-called VOD (video on demand) or NVOD (n
ear VOD) is known. FIG. 5 shows an outline of VOD. When the video signal providing side or the center side sends the video source 51 such as a movie to the subscriber side, first, the video source 51 as image information is compressed by the MPEG encoder 52. This image compression processing is based on MPEG (moving picture), which is an international standard of so-called moving image coding for storage.
A technique using correlation in the time axis direction according to the standard method of the image coding experts group) is used. The compressed signal is passed through a video multiplexer 53 to a random access memory (RAM) device 5.
4 and stored.

Upon request from the subscriber side, the compressed signal stored in the RAM device 54 is sent from the buffer 55 to the formatter 56, where it is sent to the digital network (digi).
talnetwork) It is transmitted from the exchange 57 to the subscriber side via the cable 58. On the subscriber side, a so-called receiver 59, which is a set top box, receives a compressed signal transmitted by a cable, and the received compressed signal is compressed by an MPEG decoder 60 in the receiver 59. Is expanded into an image signal, which can be viewed as an image on the television monitor 61 or the like.

Here, the compression method using the correlation of information in the time axis direction according to the above-mentioned MPEG standard method or the like can obtain a high compression rate and high image quality, so that it is also useful in effectively using the information storage space. Is.

[0005]

By the way, in the conventional image processed without compressing the image signal, when an error occurs, only a part of the image is affected, and the loss data generated at this time is a pixel. Is interpolated with the data around or the data corresponding to the same place in the front and back sections.

However, the image information obtained by the image compression method in the MPEG system is GOP (group of picture).
s), the MPEG decoder 60 in the receiver 59 on the subscriber side in FIG. 5 affects the entire GOP when a data error occurs. As an example of an image compressed by the MPEG method, taking 15 frames as a unit, if a data error occurs in one GOP, an image failure of 0.5 seconds occurs.

Further, as the location of the data error,
The following two places are mainly mentioned. One is an error in a memory device such as a RAM device, which may be a partial failure of the memory device. The other is an error in the network, and for example, in the ATM switching network, a cell drop of the information signal can be considered.

For an error in the memory device, for example, for missing data, a mirror structure is provided in the information storage space, or image information is RAID (redundant arra).
There are methods such as adding redundant memory such as ys of inexpensive disks) to enable data restoration, but these methods complicate the structure of the compressed image signal and increase the hardware. Etc. will be expensive.

Further, since video data is used as the image information, unlike computer data, the characteristics of images and visual characteristics are used to interpolate the data to such an extent that it does not matter when visualized. Therefore, the above method is inappropriate.

In the case of a cell drop error in the network, if the image is input in real time, the means of re-sending cannot be used, so an effective data restoration method cannot be found for this.

Therefore, the present invention has been made in view of the above-mentioned circumstances, and when encoding and transmitting an image signal,
A video signal code that encodes a video signal that can interpolate missing data without using an expensive data restoration method for the missing data that occurs and without having to take a means of retransmitting the missing data. It is an object of the present invention to provide an encoding device and a video signal decoding device that interpolates missing data that occurs when decoding an encoded video signal.

[0012]

A video signal coding apparatus according to the present invention comprises an even and odd screen separation unit for separating a video input signal in even and odd screens, for example, even and odd field column units, and even and odd screens. An even-numbered screen coding unit and an odd-numbered screen coding unit that independently compress the video signals divided by the screen separation unit for each screen sequence, and a compressed signal that has been compressed by these screen coding units And a video multiplexing device for

The even and odd fields may be formed for each group of pictures (GOP). Further, instead of the even and odd fields, even and odd frames may be formed.

Further, it is preferable that the signal compression performed by the even field coding unit and the odd field coding unit uses correlation in the time axis direction.

The video signal decoding apparatus according to the present invention includes an even and odd data separation unit for separating a composite video signal obtained by the video signal encoding apparatus into even and odd field GOPs, and the separated even field GOP. An even field error detecting section and an odd field error detecting section that independently detect errors in the odd field GOP signal, and an even field decoding section that independently expands the compressed signal of each field GOP after error detection, and An odd field decoding unit, an image processing unit that interpolates a defective portion in which an error is detected from the video signal obtained by each field decoding unit, and the operation of this image processing unit based on the data from the error detection unit. And a control unit for controlling.

Further, the data unit in which the composite video signal is separated by the even and odd data separating units may be a frame unit.

[0017]

According to the video signal encoding device of the present invention, the video signal is separated into even and odd field columns, and the compression processing of the video signal is performed independently for each field column by using the correlation in the time axis direction. After each field sequence is converted into each GOP, it is combined into one composite video signal for transmission to each GOP and transmitted to the video signal decoding apparatus. Further, according to the video signal decoding apparatus according to the present invention, ,
The transmitted composite video signal is again separated into even and odd field GOPs, error detection and expansion processing are independently performed for each GOP, and converted into a video signal of each field string, and expansion processing is performed in the image processing unit. A video signal can be obtained from the generated video signal based on the error information.

[0018]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of a video signal encoding device and a decoding device according to the present invention will be described below in detail with reference to the drawings.

FIG. 1 is a diagram showing a schematic configuration of a video signal encoding device according to the present invention, and FIG. 2 is a diagram showing a schematic configuration of a video signal decoding device according to the present invention.

In FIG. 1, a video signal is input from the video signal input terminal 1 and sent to the even and odd field separation section 2. In the even-numbered and odd-numbered field separation unit 2, the video signal to be transmitted is divided into field units into an even field sequence and an odd field sequence, the even field sequence is transmitted to the even field encoding unit 3, and the odd field sequence is odd field encoded. Sent to section 4 and each field string is coded independently. In addition, in this encoding,
The correlation along the time axis is used. The encoded field sequences form so-called GOPs (group of pictures), which will be described later, and are sent to the video multiplexer 5, where the individual GOPs are combined into one signal. The combined video signal combined into one is transmitted from the combined signal output terminal 6 via, for example, a digital network.

In FIG. 2, the compressed signal transmitted from the combined signal output terminal 6 of FIG. 1 via a digital network or the like is input from the combined signal input terminal 11 to the even and odd number data separation section 12. Even and odd data separation unit 12
Then, the compressed signal to be sent is again divided into GOP units,
The GOP from the even field column, that is, the even field GOP and the GOP from the odd field column, that is, the odd field GOP are separated. Even field G
The OP is sent to the even field error detector 13, and the odd field GOP is sent to the odd field error detector 14. Here, it is determined whether or not there is a missing data portion in each GOP.
OP and OP are independently detected, and the error detection result is sent to the control unit 18. Furthermore, each GOP is
It is sent to the even field decoding unit 15 or the odd field decoding unit 16 corresponding to the OP and is decoded independently, and becomes the video signal of the even field sequence and the odd field sequence. The decoded video signal of each field sequence is sent to the image processing unit 17. Image processing unit 17
On the basis of error information from the control unit, when there is a missing data portion in each field signal to be sent, the operation of discarding the video signal of the field sequence is performed.
Furthermore, when either the video signal of the even field row or the video signal of the odd field row is discarded, the video signal close to the original signal is obtained by performing the interpolation processing based on the video signal of the remaining field row. . The obtained video signal is sent from the video signal output terminal 19 to a television monitor or the like and displayed on the monitor screen.

The operation of the decoding process may be ON / OFF controlled according to the error detection output.

FIG. 3 is a diagram showing an outline of the GOP.

In FIG. 3, GOP consists of 15 fields, which are classified into I screen, P screen, and B screen, respectively. The I screen represents an intra-field coded screen, the P screen represents an inter-frame predictive coded screen, and the B screen represents a bidirectional predictive coded screen. The P screen is predicted from the I or P screen located in front, and the B screen is bidirectionally predicted from the I or P screen located in the front or rear.

FIG. 4 is a diagram showing an outline of the flow of a video signal processed by the video signal encoding device and the decoding device according to the present invention.

In FIG. 4, the input original image frame sequence 31 is separated into an even field sequence 32 and an odd field sequence 33 by the even and odd field separation unit 2.
Subsequently, the signal of each field sequence is encoded by the even field coding unit 3 and the even field coding unit 4 corresponding to each field, and the even field GOP 34
And an odd field GOP35. These two GOPs are combined into a single signal by a video multiplexer or the like, and the combined composite video signal is stored in a memory device 36 in the video multiplexer. In response to a request from the user, the necessary composite video signal is supplied to the user through the digital network 37. The supplied composite video signal is again separated into an even field GOP and an odd field GOP by the even and odd data separating section 12, and these GOPs are respectively an even field error detecting and decoding section 38 and an odd field GOP. The error detection / decoding unit 39 independently performs error detection and decoding processing to form an even field sequence 40 and an odd field sequence 41. These video signals are processed by the image processing unit according to the error information detected by the error detection unit or the like. FIG. 4 shows the case where an error is detected in the even field column, and therefore, in the image processing unit,
The signal of the even field sequence 40 is discarded, and the odd field sequence 41 is interpolated to obtain the interpolated image frame sequence 4
Get 2.

With the above-described structure, the video signal to be transmitted is separated into two field strings, each of which is independently encoded, and each field string is converted into a GOP-unit signal. Are combined into one signal and transmitted, and this video signal is received, separated into GOPs again, independently decoded, and converted into the video signal of each field sequence, and then either field is temporarily Even if an error is detected in the video signal of the column and the image is damaged, the original video signal can be obtained by interpolating the missing data based on the video signal of the other field column.

Even if there is some visual difference due to the interpolation processing of missing data, one GOP consisting of 15 fields has a short time of 0.5 seconds.
When applied to an application that assumes VHS level image quality such as VOD, there is no problem. Also,
According to this method, it is possible to provide a video signal encoding device and a decoding device which do not require an extra information storage space and a processing device in a memory device and can cope with accidents such as cell loss in a network.

[0029]

As described above, according to the video signal coding apparatus of the present invention, the video signal is separated into the even field and the odd field, thereby enhancing the correlation between the data units. , By encoding each independently, even if, for example, an accident occurs in a network communication network or the like and missing data occurs, a special processing device is not used or the transmitted video information does not have a special data structure. ,
It is possible to transmit a signal without losing data.

As described above, according to the video signal decoding apparatus of the present invention, when the video signal is compressed, the video signal is an even field column video signal and an odd field column video signal. , And each of them is independently encoded, and the combined signal is separated into an even field and an odd field again, and when they are independently decoded, an error is detected and an error detection output is performed. Even if an accident occurs in a network communication network and missing data is generated by image processing by using image processing, decoding is possible without using a special processing device or providing the transmitted video information with a special data structure. Missing data can be interpolated during processing.

[Brief description of drawings]

FIG. 1 is a block diagram showing a schematic configuration of an embodiment of a video signal encoding device according to the present invention.

FIG. 2 is a block diagram showing a schematic configuration of an embodiment of a video signal decoding device according to the present invention.

FIG. 3 is a diagram illustrating a GOP (group of pictures).

FIG. 4 is a diagram illustrating the flow of video signals handled by the video signal encoding device and the decoding device according to the present embodiment.

FIG. 5 is a diagram illustrating VOD (video on demand).

[Explanation of symbols]

 1 Video Signal Input Terminal 2 Even and Odd Field Separation Section 3 Even Field Coding Section 4 Odd Field Coding Section 5 Video Multiplexing Device 6 Composite Signal Output Terminal 11 Composite Signal Input Terminal 12 Even and Odd Data Separation Section 13 Even Field Error Detecting unit 14 Odd field error detecting unit 15 Even field decoding unit 16 Odd field decoding unit 17 Image processing unit 18 Control unit 19 Video signal output terminal 31 Original image frame sequence 32 Even field sequence 33 Odd field sequence 34 Even field GOP 35 Odd field GOP 36 Memory device 37 Digital network 40 Even field sequence 41 Odd field sequence 42 Interpolated image frame sequence

Claims (6)

[Claims] 1. A coding device for coding a video input signal, comprising: a screen separation means for separating the video input signal into an even screen and an odd screen in screen units; and an even and an odd screen by the screen separation means. Video signal encoding, characterized in that it has signal compression means for independently compressing the video signals separated on the screen, and compressed signal synthesizing means for synthesizing the compressed video signals obtained by the signal compression means. apparatus. 2. The video signal encoding device according to claim 1, wherein the screen separation means separates the video input signal into an even field and an odd field. 3. The signal compression by the signal compression means is
The video signal encoding apparatus according to claim 1, wherein correlation in the time axis direction is used. 4. A video signal coding apparatus separates a video input signal into an even field and an odd field, and decodes a composite video signal obtained by combining the signals obtained by independently encoding the signals. A video signal decoding device for performing error correction or correction of the video signal, wherein the composite video signal separating means for separating the composite video signal in screen units, and the even and odd screens obtained by the composite video signal separating means Error detecting means for independently detecting an error in the video signal, the signal expanding means for independently expanding the video signal after the error detection by the error detecting means in an even screen and an odd screen, and the signal expanding means. Video signal decoding, characterized in that it has image signal processing means for processing the image signal obtained in 1. and control means for controlling the image signal processing means. Location. 5. The video signal decoding apparatus according to claim 4, wherein the composite video signal separating means separates the composite video signal into an even field and an odd field. 6. The video signal decoding apparatus according to claim 4, wherein the control means operates based on a signal from the error detection means.