Motion video signal processing for recording or reproducing – Local trick play processing – With randomly accessible medium
Reexamination Certificate
1998-07-17
2001-09-11
Garber, Wendy R. (Department: 2615)
Motion video signal processing for recording or reproducing
Local trick play processing
With randomly accessible medium
C386S349000, C714S746000
Reexamination Certificate
active
06289173
ABSTRACT:
TECHNICAL FIELD
The present invention relates to a recording and reproducing apparatus for recording digital video signals subjected to compression coding for image coding.
BACKGROUND ART
In recent years, a variety of digital instruments have been developed as digital signal processing techniques are improved. Small-sized digital video recorders for recording digital video signals in the form of compressed codes have been realized lately. Among various methods that have been proposed for the compression coding of digital video signals, two-dimensional discrete cosine transform (DCT) coding, which is one of the methods for frequency conversion of digital video signals in the unit of a block, has become the mainstream as used in JPEG and MPEG. In this method, digital video signals are divided into a plurality of coding blocks and discrete cosine transform is performed for each coding block, and further the resulting coefficient data are subjected to variable-length coding for each block, to compress data.
However, in such a recording and reproducing apparatus using compression coding, if an error has occurred to data during the recording or reproducing process, variable-length codes in a certain range subsequent to the error occurrence point could not be decoded. Also, even if the variable-length decoding is refreshed by detecting a break in the variable-length code by means of refresh codes or the like, the variable-length codes themselves over a range from the occurrence of an error to the refreshing could not be decoded. In such a case, the image would drop out in the unit of the block on the reproduction screen, and the dropping out would succeed over several blocks.
In order to solve this problem in the compression coding, in the recording and reproducing apparatus with image coding described in Japanese Patent Laid-Open Publication 7-298194/1995 by the present inventors, errors are corrected by using coefficient data of neighboring coding blocks having high spatial correlation. In this error correction, for recording, coefficient data within one coding block are mixed with coefficient data of a boundary portion within a neighboring coding block and subjected to shuffling. In the shuffling, it has also been proposed to divide neighboring coding blocks so that their boundary portion is shared therebetween. Then, the discrete cosine transform is performed with respect to coding blocks constituting the same region on the screen, the resulting coefficient data are classified into groups, and the groups are recorded in separate regions on the recording medium. In this apparatus, even if an error occurs during recording or reproducing process, there will never occur any collective disappearance of associated coefficient data. Further, even if coefficient data has disappeared upon occurrence of an error, correct data can be restored more accurately by using the correlation of coefficient data. Therefore, the stability of recording and reproducing operations can be greatly enhanced without causing the image to disappear in blocks.
However, even in this recording and reproducing apparatus with image coding, if a plurality of errors of coefficient data have occurred within the same coding block, correct restoration by error correction cannot be done. It is desirable that an error can be corrected so that image is not dropped out in such a situation as mentioned above.
SUMMARY OF THE INVENTION
An object of the present invnetion is to provide a recording a reproducing apparatus with image coding in which an image will never be dropped out in blocks even if a plurality of errors occur when digital video signals are recorded or reproduced with compression coding.
In a recording and reproducing apparatus with image coding according to the present invention, for recording process, overlap blocking means divides input digital video signals into data in two-dimensional coding blocks each having first and second axes and comprises a plurality of pixel data, in such a way that neighboring coding blocks are overlapped with each other so as to share at least one boundary pixel in a first-axis (for example, horizontal) direction. Orthogonal transform means performs two-dimensional frequency conversion (for example, discrete cosine transform) for pixel data of coding blocks obtained by the overlap blocking means to generate coefficient data of individual frequencies. Shuffling means divides the coefficient data obtained by the orthogonal transform means into a plurality of coefficient groups so that the coefficient groups do not have the same frequency component of the second (for example, vertical) axis at least in a frequency region in the coding blocks corresponding to low frequency components of the first (for example, horizontal) axis. Recording means locates recording data so that the plurality of coefficient groups in the same coding block obtained by the shuffling means are recorded so as to be dispersed into separate regions on a recording medium, and it adds error correction codes for the coefficient data.
For a reproducing process, reproduction decoding means reproduces recording data from the recording medium and decodes the coefficient data of the individual coefficient groups, while it detects the presence or absence of an error of coefficient data. Deshuffling means integrates coefficient data outputted from the reproduction decoding means into coding blocks for recording. When an error has been detected by the reproduction decoding means, error-correction control means outputs an error-correction control signal based on the number of errors within a coding block and the number of the order of the two-dimensional frequencies of coefficient data to which errors have occurred, and error correcting means corrects the errors of the coefficient data by an error correcting method selected by an error-correction control signal of the error correction control means with respect to the coding block in which errors have been detected. block in which errors have been detected.
Inverse orthogonal transform means decodes pixel data by frequency inverse conversion of the coefficient data that have been error corrected by the error correcting means, and relocating means generates a digital video signal by relocating pixel data to the coding blocks for recording. When an error has occurred, such a possibility can be prevented that all the coefficient data within a block may be dropped out so as to lack in blocks during reproducing process. Further, the stability of reproduced images can be enhanced by performing error correction with coefficient data of neighboring blocks having high spatial correlation.
Preferably, the recording means locates recording data, so that coefficient groups belonging to the same coding block are not recorded to the same recording track on the recording medium. Upon occurrence of an error, such a possibility can be prevented that all the coefficient data within the block may be dropped out so as to lack in the unit of block during reproducing process.
Preferably, the recording and reproducing apparatus with image coding further comprises a rotary cylinder for helically scanning the recording medium, and signal recording means having a plurality of recording heads attached to the rotary cylinder. The recording means outputs recording data to the signal recording means so that coefficient groups including coefficient data corresponding to low frequency components of the first (for example, horizontal) axis within at least two overlapped coding blocks in the overlap relationship are not recorded to the same recording head. In a different way, the recording means outputs recording data so that coefficient groups including coefficient data corresponding to the same horizontal frequency component in two overlapped coding blocks in the overlap relationship are not recorded to the same recording head. When an error has occurred, such a possibility can be prevented that all the coefficient data within the block may dropped out so as to lack in blocks during reproducing process.
Bannai Tatsushi
Miyashita Mitsuhiro
Chieu Po-lin
Garber Wendy R.
Matsushita Electric - Industrial Co., Ltd.
Wenderoth , Lind & Ponack, L.L.P.
LandOfFree
ECL in a recording and reproduction device in which... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with ECL in a recording and reproduction device in which..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and ECL in a recording and reproduction device in which... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2451617