Image analysis – Image transformation or preprocessing – Combining image portions
Patent
1997-11-12
1999-11-30
Tadayon, Bijan
Image analysis
Image transformation or preprocessing
Combining image portions
382282, G06K 900
Patent
active
059956798
DESCRIPTION:
BRIEF SUMMARY
BACKGROUND OF THE INVENTION
In the encoding of image sequences that represent a combination of computer image parts, for example image parts that are constructed from computer graphics, and pixel-based image parts, for example image parts that are registered by a camera, it is known to separately encode and transmit the input data of the computer graphics (script files with associated data) and the pixel-based image parts. It is only at the receiver that the graphics models are then rendered and mixed with the pixel-based image parts. Since the mixing is not a binary operation, the information about the ratio in which the mixing should be carried out must be additionally transmitted for each pixel. This information is stored in a key signal and is acquired by a color segmenting from the camera signal in traditional television (F. Fechter et al., "Signal processing for a digital HDTV chromakey mixer", Signal Processing: Image Communications 5, 1993, pp. 417-423 and T. Sprague, "Merging Digital Technology into an Analog World", SMPTE Journal, February 1994, Vol. 103, No. 2, 1994, pp. 100-104).
Methods are known for transmission that separately encode computer image parts and pixel-based image parts in a transmission unit, together with what is referred to as a key signal that indicates the ratio in which a pixel of a computer image part is to be respectively mixed with the corresponding pixel of the pixel-based image part. In blue box exposures, for example, the key signal is acquired from the pixel-based image parts by a color segmenting. The intensity of the blue coloration thereby indicates the later mixing ratio in which the pixel information of the pixel of the computer image part is mixed with the corresponding pixel information of the pixel-based image part in the receiver unit. The mixing of the two pixel information in the receiver unit is not a binary mixing.
Since the mixing is not binary, much information for the respectively mixing ratio must be transmitted for every dot with the assistance of the key signal. The key signal thus requires a considerable amount of transmission capacity. However, the information of the key signal has no additional informational content for the image itself but merely the information about the ratio in which the images should be mixed together in the receiver unit.
SUMMARY OF THE INVENTION
The invention is based on the problem of already partially mixing image sequences composed of computer image parts and pixel-based image parts in a suitable way and encoding them and thereby eliminating the transmission of the key signal that is normally additionally required.
In general terms the present invention is a method for encoding and decoding image sequences using computer image parts and pixel-based image parts on the basis of a key signal. In a transmission unit, pixel information of a pixel of the computer image part is mixed dependent on the key signal with pixel information of a corresponding pixel of the pixel-based image part to form a partially mixed image sequence. The partially mixed image sequence and the computer image part are transmitted from the transmission unit to a receiver unit. Only completely covering regions of the computer image part are rendered in the receiver unit. The completely covering regions of the computer image part are inserted into predetermined regions of the partially mixed image sequence.
Advantageous developments of the present invention are as follows.
Regions of the partially mixed image sequence that are completely covered by pixels of the computer image parts are blanked out before the encoding.
Regions in the partially mixed image sequence having the block size of the encoder that are completely covered by the computer image parts have a constant luminance value assigned to them.
The inventive method discloses a possibility of modifying the pixel-based image parts of image sequences dependent on the key signal before the encoding in order to be subsequently able to efficiently compress them with block-based encoders (for ex
REFERENCES:
patent: 5113251 (1992-05-01), Ichiyanagi et al.
patent: 5539866 (1996-07-01), Banton et al.
patent: 5805228 (1998-09-01), Proctor et al.
patent: 5892847 (1997-04-01), Johnson
Fernseh-und Kino-Technik vol. 47, No. 1, S. Hartwig et al, (1993), (Tutorial) Digitale Bildcodierung, pp. 33-42.
Plenary, Special Audio. Underwater Acoustics, VLSI, Neural Networks, (1993) IEEE International Conference on Acoustics, Speech, and Signal Processing, pp. 99-102. (No Author).
Signal Processing: Image Communications, vol. 5, (1993), Frank Fechter et al, Signal Processing for a Digital HDTV Chromakey Mixer, pp. 417-423.
SMPTE Journal, vol. 103, No. 2, Feb. 1994, T. Sprague, Merging Digital Technology into an Analog World, pp. 100-104.
"Coding of Combined Natural & Computer Rendered Image Sequences", Thomas Riegel, SPIE vol. 2451/207-211, pp. 207-211, Munich, Germany, 1995, 0-8194-1802-1/95/.
Siemens Aktiengesellschaft
Tadayon Bijan
LandOfFree
Process for coding image sequences in a transmitter unit does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Process for coding image sequences in a transmitter unit, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Process for coding image sequences in a transmitter unit will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-1684228