Pulse or digital communications – Bandwidth reduction or expansion – Television or motion video signal
Reexamination Certificate
1999-02-24
2003-06-03
Philippe, Gims S. (Department: 2613)
Pulse or digital communications
Bandwidth reduction or expansion
Television or motion video signal
C375S240100, C348S416100, C382S251000
Reexamination Certificate
active
06574274
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a picture signal processing system, a decoder, a picture signal processing method, and a decoding method suitable for an application that requires high picture quality. In addition, the present invention relates to a picture signal decoding method, a decoding apparatus, a picture signal encoding method, an encoding apparatus, a picture signal processing method, and a processing system for decoding a picture signal that has been compression-encoded by for example MPEG encoding method and encoding the decoded signal.
2. Description of the Related Art
In recent years, the MPEG (Moving Picture Experts Group) encoding method, which is an inter-picture compression encoding method, has been widely used. With the compression-encoding method such as the MPEG encoding method, a record medium can be effectively used. In an editing process for a picture signal that has been encoded by the MPEG encoding method, when a picture signal that has been encoded is decoded and then the decoded picture is re-encoded, the picture quality sometimes deteriorates more than that of a normal encoding process for a video signal. As a cause of such deterioration, encoding parameters such as a picture type and a moving vector of a picture signal encoded by the MPEG encoding method do not match those of a picture signal re-encoded by the MPEG encoding method.
Next, with reference to
FIGS. 9A
,
9
B, and
9
C, the case where a picture type of a picture signal that is encoded does not match that of a picture signal that is re-encoded will be described.
FIG. 9A
shows an example of picture types of input decoded pictures of one GOP (Group Of Pictures) (n=15) where the input decoded pictures are re-encoded. When the phases of picture types are locked as shown in
FIG. 9B
, as a reference picture for a re-encoding process, an I picture of the input decoded picture shown in
FIG. 9A
is used as is.
On the other hand, when the phases of picture types are not locked as in the case shown in
FIG. 9C
, as with the third picture, a B picture whose picture quality largely deteriorates is used as a reference picture. As a result, the accuracy of the re-encoding process deteriorates. Thus, the picture quality largely deteriorates.
Likewise, when another encoding parameter such as a moving vector of a picture signal that is encoded does not match that of a picture signal that is re-encoded, the predictive accuracy deteriorates. Thus, the accuracy of the re-encoding process deteriorates.
On the other hand, when all encoding parameters including a picture type and a moving vector of a picture signal that is encoded are matched with those of a picture signal that is re-encoded, the picture quality hardly deteriorates. However, as the number of encoding parameters increases, the amount of information to be processed increases. Consequently, it is not practical to perform a process for matching all encoding parameters of a picture signal that is encoded with those of a picture signal that is re-encoded. It is known that when a picture type and a moving vector of a picture signal that is encoded are matched with those of a picture signal that is re-encoded, the deterioration of the picture quality can be remarkably suppressed.
However, when high picture quality is required, even if a picture type and a moving vector of a picture signal that is encoded are matched with those of a picture signal that is re-encoded, the deterioration of the picture quality cannot be sufficiently suppressed.
A moving vector represents information of motion of each macro block. Thus, the amount of information of a moving vector is relatively large. Consequently, from a view point of the improvement of the re-encoding process, an encoding parameter that has a smaller amount of information than that of a moving vector and that effectively suppresses the deterioration of the picture quality is required in the case that the structure for matching a moving vector of a picture signal that is encoded is matched with that of a picture signal that is re-encoded cannot be used.
In a system using the MPEG encoding method, an MPEG encoded stream is decoded. The resultant picture signal is converted into a format of a transmission picture signal such as an NTS format. The resultant picture signal is transmitted through a digital VCR or the like and re-encoded by the MPEG encoding method. In another editing system, a picture signal that has been reproduced from a record medium and decoded is combined with an external picture signal. The resultant picture signal is encoded by the MPEG encoding method. The resultant stream is recorded on a record medium. In a dubbing process for recording a video signal recorded on a first record medium to a second record medium by a recorder, a signal of the first record medium is decoded and then re-encoded.
In such examples, when a decoded picture signal is re-encoded, it is preferred to match a picture encoding type of a stream that is re-encoded (an output signal of an encoder) with that of a stream that is encoded (an input signal of a decoder) so as to improve the picture quality of a stream that has been re-encoded and then decoded. In the MPEG encoding method, there are three picture types I, P, and B.
In an I picture (Intra-coded picture), when a picture signal is encoded, information of only one picture is used. Thus, when an encoded picture signal is decoded, information of only the I picture is used. In a P picture (Predictive-coded picture), as a predictive picture (a reference picture for obtaining a difference with the current P picture), an I picture or another P picture that has been decoded is temporally followed by the current P picture. The difference between the current P picture and a motion-compensated predictive picture is encoded for each macro block. Alternatively, the current P picture is encoded for each macro block without obtaining the difference of such pictures. One of those methods is selected whichever higher efficiency is obtained. In a B picture (Bidirectionally predictive-coded picture), as predictive pictures (reference pictures for obtaining a difference with the current B picture), three types of reference pictures are used. The first type reference picture is an I picture or a P picture that has been decoded and that is temporally followed by the current B picture. The second type reference picture is an I picture or a P picture that has been decoded and that is temporally preceded by the current B picture. The third type reference picture is an interpolated picture of the first type reference picture and the second type reference picture. The difference between the current B picture and each of the three type reference pictures that have been motion-compensated is encoded for each macro block. Alternatively, the current B picture is encoded for each macro block without obtaining such a difference. One of those methods is selected whichever higher efficiency is obtained.
Thus, there are a frame intra-coded macro block, a forward inter-frame predictive macro frame (a future macro block is predicted with a past macro block), a backward inter-frame predictive macro block (a past macro block is predicted with a future macro block), and an interpolative macro block (a current macro block is predicted with both a future macro block and a past macro block). All macro blocks in an I picture are intra-frame coded macro blocks. A P picture contains intra-frame coded macro blocks and forward inter-frame predictive macro blocks. A B picture contains the above-described four types of macro blocks.
Generally, the picture quality of a decoded picture of an I picture or a P picture is worse than the picture quality of a decoded picture of a B picture. When a picture is re-encoded, if a decoded picture of a B picture is substituted with an I picture or a P picture and then a B picture is generated, the picture quality of the resultant B picture deteriorates. Thus, when the picture encoding type of a picture that
Igi Nobuhiro
Kato Motoki
Obata Koji
Frommer William S.
Frommer & Lawrence & Haug LLP
Philippe Gims S.
Savit Glenn F.
Sony Corporation
LandOfFree
Picture signal processing system, decoder, picture signal... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Picture signal processing system, decoder, picture signal..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Picture signal processing system, decoder, picture signal... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3115868