JP3200515B2 - Processing method for encoded and decoded images - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a multi-step process of decoding an encoded image, subjecting the obtained decoded image to processing such as editing, and re-encoding and transmitting / accumulating the obtained image. In the case of repetition, by matching various coding parameters generated during editing or multi-stage connection processing, as a result, a coded decoded image that can minimize image quality deterioration even if coding and decoding are repeated In the processing method.

[0002]

2. Description of the Related Art Consider a case where a decoded image obtained by high-efficiency encoding and decoding is re-encoded.

[0003] In this case, depending on the contents of the processing, it is not always possible to execute a code string obtained by encoding. For example, when performing editing processing such as moving a part of the screen to another part, extracting a certain time part from one sequence and inserting it into another sequence, or creating a slow motion video , Once the code string is decoded and PCM (Pulse Code Mod
ulation) It becomes necessary to execute processing after obtaining a decoded image of the signal.

[0004] Considering a case where an image obtained by encoding and decoding is subjected to editing processing and the like, and then transmitted and stored with high-efficiency encoding again, encoding and decoding are repeated a plurality of times. A situation of multistage connection will occur.

[0005] FIG. 4 is a diagram showing an example of a configuration of a multi-stage connection of encoding and decoding according to a conventional method when editing an encoded and decoded image is involved. First, the PCM digital image signal 82 input from the input terminal 81 is input to the first encoding unit 8.
3, high-efficiency encoding is performed, and a first code string 84 is obtained. After being transmitted or stored, the first code sequence 84 is decoded by the first decoding unit 85, and a first decoded image 86 of the PCM signal is obtained. An editing process is performed on the obtained decoded image in the image editing unit 87, and an edited image 88 is obtained. Next, the edited image 88 is again input to the second encoding unit 89 for transmission or storage, and is subjected to high-efficiency encoding to obtain a second code sequence 90. After being transmitted or stored, the second code string 90 is decoded by the second decoding unit 91 to obtain a second decoded image 92 of the PCM signal.

[0006]

As described above, as described above,
Considering the case where an image obtained by encoding / decoding is subjected to processing such as editing, and then is subjected to high-efficiency encoding and transmission / storage, etc., encoding / decoding is repeated a plurality of times.
That is, a situation of multi-stage connection occurs. In multi-stage connection for encoding / decoding, generally, the image quality of a decoded image obtained gradually decreases as the number of connection stages increases. The degree of image quality degradation differs depending on the connection mode, but generally, the degree of image quality degradation increases when there is a difference in encoding parameters in encoding at each stage.

In the conventional multi-stage connection method, only the decoded image obtained as a PCM signal is handled regardless of the encoding parameters when the decoded image to be processed is encoded. Depending on the type, when the image after editing is re-encoded, there is a possibility that a mismatch occurs with the encoding parameter used when the image before editing was obtained.

For example, MPEG (Motion Picture Exp)
In the ert Group) coding method, as shown in FIG. 5, the coding type of a frame is defined as an I picture to be coded only within the frame, a P picture to be coded using a temporally past frame, And B-pictures to be coded using both past and future frames. In this case, when encoding and decoding are connected in multiple stages, the picture quality can be minimized by matching the picture types. However, after editing using only the decoded image that has been converted into a PCM signal, the editing is performed again. In the case of encoding, the picture type of the frame cannot be referred to at the time of editing processing or re-encoding. Therefore, since the picture type at the time of re-encoding does not always match the picture type used at the time of the previous encoding, there is a problem that the degree of image quality degradation becomes large when multiple connections are made. there were.

[0009]

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problem, and has been made in order to process a decoded image signal obtained by decoding a code sequence obtained by performing high-efficiency coding. In this case, an encoding parameter used for encoding is extracted from a code string, and processing such as editing or re-encoding of a decoded image signal is performed with reference to the extracted parameter.

FIG. 1 is a block diagram showing the principle of the present invention. In FIG. 1, 1 is an input terminal, 2 is a code sequence, 3 is an image decoding unit, 4 is a decoded image, 5 is an encoding parameter extracting unit, 6 is an encoding parameter, 7 is an image processing unit, and 8 is image processing. Signal 9 indicates an output terminal.

A code sequence 2 input from an input terminal 1 is decoded by an image decoding unit 3 to obtain a decoded image 4 of a PCM signal.
Is obtained. On the other hand, a coding parameter extraction unit 5 extracts a coding parameter 6 used when obtaining the code string 2 from the code string 2. The image processing unit 7 edits or processes the decoded image 4 of the PCM signal while referring to the encoding parameter 6, and outputs the edited or processed image-processed signal 8 to the output terminal 9. In the present invention
Is highly efficient coded, especially according to multiple picture types.
Are obtained by the same coding method in the first code string.
When inserting the second code string, the image processing unit 7
The picture type structure of the second code string to be inserted is the first
Inserted to match the picture type structure of the code string of
A decoded image obtained by adjusting the start frame and decoding the first code string.
Decoding the second code string from the insertion start frame into the image signal
And inserts the decoded image signal into the first decoded image signal.
The same picture type as the picture type structure of the code string of
Encode by structure.

[0012]

According to the method as described above, when processing such as editing / encoding is performed on an image in which a code string obtained by encoding an image signal with high efficiency is once decoded into a PCM signal. It is possible to perform the decoding while referring to the encoding parameters when the decoded image to be encoded is encoded. Therefore, in consideration of re-encoding an image obtained by performing processing such as editing on a decoded image, processing such as editing can be performed so that the coding parameters match.

[0013]

Examples of the present invention will be described below. In this embodiment, it is assumed that an image encoded and decoded by the MPEG system is edited and transmitted or stored again by the MPEG system. For editing, it is assumed that a part is extracted from one sequence and incorporated into a part of another sequence.

In the MPEG coding method, the coding type of each frame is defined as an I picture to be coded only within that frame, a P picture to be coded using a temporally past frame, a temporally past and future Encoding is performed by dividing into three types of B pictures to be encoded using both frames. When MPEG encoding / decoding is connected in multiple stages,
It is known that image quality degradation can be minimized by matching picture types.

As shown in FIG. 2, there are two sequences S1 and S2 as an image code sequence, and the picture type structure in the first stage of encoding / decoding is a P picture every two frames and a P picture every 6 frames. Is an I picture. Consider editing such that the portion from the m1th frame to the m2th frame of the decoded image sequence S1 'is overwritten to the portion starting from the n1th frame of the decoded image sequence S2'. In this case, as shown in FIG. 2, the insertion start frame n1 is set so as to match the picture type structure of the m1 to m2 frames of the sequence S1.
Is adjusted, and overwriting is performed with the adjusted insertion start frame n1 'at the top. That is, overwriting is performed so that an I picture is an I picture, a P picture is a P picture, and a B picture is a B picture, so that an edited image sequence S3 is obtained.

In this manner, when the sequence S3 is re-encoded, the picture type used in the first stage for each frame is referred to by referring to the picture type extracted for the sequence S2 '. The same type as the type can be applied. In the first stage, a frame encoded with an I picture is an I picture, a frame encoded with a P picture is a P picture, and a frame encoded with a B picture is a B picture.
It becomes possible to encode with a picture.

FIG. 3 shows a block configuration of an embodiment of the present invention. The image signal S1 input from the first input terminal 11
Are encoded in the first MPEG encoding unit 12,
The obtained first code string 13 is transmitted or stored, and is decoded by the first MPEG decoding unit 14, and the first decoding P
S1 'is obtained as the CM image 15.

On the other hand, the image signal S2 inputted from the second input terminal 16 is encoded in a second MPEG encoding section 17, and the obtained second code sequence 18 is transmitted or stored, and Is decoded by the MPEG decoding unit 19 of
S2 'is obtained as the second decoded PCM image 20.

The MPEG decoders 14 and 19 decode the code string, and at the same time, extract the picture type (I picture, P picture, B picture) for each frame from each image signal, and store them in the coding parameter memory. 21 and 22. In the inserted image extracting unit 23, frames m1 to m2 at a predetermined time from the first decoded PCM image 15 are extracted as an inserted image 24, and a picture type 25 corresponding to each frame of the extracted image is encoded. It is read from the parameter memory 21 and sent to the insertion start frame adjustment unit 26. The insertion start frame adjusting unit 26 compares the picture type 25 for the inserted image 24 with the picture type 27 for the second decoded image, and matches the picture type at the position closest in time to the specified insertion start frame n1. Adjust the insertion start frame as follows.

In the image insertion unit 28, the insertion image 24 is inserted starting from the insertion start frame n1 'determined by the insertion start frame adjustment unit 26,
3 is obtained. To encode the edited image S3 again, the encoding parameter memory 22 for the image S2 'is used.
Is read out from the first MPEG encoding unit 30 and encoded by the third MPEG encoding unit 30,
In addition, the picture type can be matched with that of the second MPEG encoding unit 19.

In the above embodiment, the case where the editing is performed with reference to the picture type of the frame has been described.
Adjusting the position of the motion compensation block or the DCT block so as not to shift can be considered in exactly the same manner.

[0022]

As described above, according to the present invention, when a code string obtained by encoding a video signal with high efficiency is once decoded into a PCM signal, and the image is subjected to processing such as editing and coding. In addition, the decoding can be performed by referring to the encoding parameters when the decoded image to be processed is encoded. Accordingly, it is possible to match the encoding parameters when encoding / decoding is connected in multiple stages, and as a result, to minimize the deterioration in image quality when processing such as encoding / decoding / editing is connected in multiple stages. Leads to things.

[Brief description of the drawings]

FIG. 1 is a principle block diagram of the present invention.

FIG. 2 is a diagram illustrating a processing operation in an embodiment of the present invention.

FIG. 3 is a block diagram of an embodiment of the present invention.

FIG. 4 is a diagram illustrating a configuration example of a multi-stage connection of encoding and decoding according to a conventional method.

FIG. 5 is a diagram showing a picture type in MPEG.

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 input terminal 2 input code sequence 3 image decoding unit 4 PCM decoded image 5 encoding parameter extraction unit 6 encoding parameter 7 image processing unit 8 image processed signal 9 output terminal 11 first input terminal 12 first MPEG code Encoding unit 13 first code sequence 14 first MPEG decoding unit 15 first decoded PCM image 16 second input terminal 17 second MPEG encoding unit 18 second code sequence 19 second MPEG decoding Unit 20 second decoded PCM image 21 first encoding parameter memory 22 second encoding parameter memory 23 inserted image extraction unit 24 inserted image 25 picture type for inserted image 26 insertion start frame adjustment unit 27 second decoded image 28 picture insertion unit 29 edited image 30 third MPEG encoding unit

────────────────────────────────────────────────── ─── Continuation of the front page (56) References JP-A-5-308525 (JP, A) JP-A-6-284414 (JP, A) JP-A-7-154802 (JP, A) 3 persons, "Study of coding control method in video re-encoding", 8th Symposium on Image Coding (PCSJ93), IEICE, October 1993, p. 27 (58) Fields surveyed (Int.Cl. ⁷ , DB name) H04N ^7/ 24-7/68 H04N 1/41-1/419 H04N 5/91-5/956

Claims (2)

(57) [Claims] 1. A digital video signal, such as a television signal, which is highly efficient encoded according to a plurality of picture types .
In the code string of No. 1, the
2 is a method of processing an encoded / decoded image in which
I, a process for decoding the first code string, the method comprising: decoding the second code string, the picture type structure of the second code sequence the inserted said
According to the picture type structure of the first code string,
Adjusting the insertion start frame and adding the insertion frame to a decoded image signal obtained by decoding the first code string.
A decoded image obtained by decoding the second code string from a start frame
A process of inserting a signal, picture relevant decoded image signal after the insertion of the first code string
Encoding with the same picture type structure as the key type structure
And a method for processing an encoded / decoded image. 2. The processing of the encoded and decoded image according to claim 1.
In the method, from the first and second code string, the code string to extract the coding parameters used in encoding for obtaining the extracted coding parameters recorded with decoded image signal, A method for processing a coded decoded image, comprising reading and referring to recorded coding parameters when processing a decoded image signal.