Facsimile and static presentation processing – Facsimile – Specific signal processing circuitry
Patent
1989-04-21
1990-06-05
Britton, Howard W.
Facsimile and static presentation processing
Facsimile
Specific signal processing circuitry
358135, 358136, A04N 7133
Patent
active
049318691
DESCRIPTION:
BRIEF SUMMARY
BACKGROUND OF THE INVENTION
The invention relates to a method for reducing the data rate of digitized images, the image data of which are first transformed block by block, subsequently coded and finally stored temporarily before they are output, each image data block being assigned on the basis of its own image definition statistics to a classification factor from a collection of predetermined classes corresponding to different local resolutions and the image data blocks being coded according to the classes assigned to them, and a control variable being derived from the extent of the temporarily stored data, with which control variable the data extent of image data blocks is changed.
The processing and transmission of digitized images--the description below relates to television images as an example, whereby, however, also videotelephone, images for videoconferences etc. are included--offer several advantages in comparison with analog methods. For instance, during joint processing and transmission, several different signals may be better combined and separated, malfunctions can be more easily rectified and the signals can be processed and amplified intermediately without loss of quality even on a longish transmission route.
These advantages are offset by the disadvantage of the considerably higher transmission capacity in comparison with analog methods, which is understood to include, inter alia, a high bandwidth, bit rate and processing speed. The available transmission capacity, be it free frequency ranges in the case of radio transmission or free lines of the telecommunication network in the case of wire-based transmission, is however, limited, and would greatly restrict the number of transmission channels for digitized television images and be very costly for the users.
Application examples such as videotelephones or videoconferences require, on the other hand, an extensive network of transmission channels and low operating costs, so that enough subscribers see the acquisition as useful and justify setting up a network of this kind.
In order to fulfil these marginal conditions, methods have been developed for data reduction in which the data extent is controlled depending on the local image statistics. Since the average data extent is less than the maximum required for representing structures of high definition, a data reduction is thus achieved.
In a known method of this type, as is described in the research report "Kombinierte Quellen- und Kanalkodierung in adaptiven Transformations-Kodierungssystemen" (Combined Source and Channel Coding in Adaptive Transformation Coding Systems), Fortschrittsberichte VDI, No. 52, 1986, pages 74-117, the image data of digitized television images are first transformed block by block in a transformer circuit, subsequently coded in a coder and finally temporarily stored in a buffer memory before a serial output.
The coding is not the same for each image data block. Rather, each image data block is assigned a class code or a classification factor determined from its own image statistics and the image data blocks are then differently coded according to this assignment. In doing so, the coding is such that for the total image an extent of image data is obtained which, with temporary storage in the buffer memory, leads to the output of a data rate lying below a predetermined limit value.
This method functions properly as long as television images of originals of an average definition (i.e., richness in detail and contrasts) are to be transmitted. In the case of originals of high definition, on the other hand, there is the danger that the buffer memory will overflow, which would lead to far-reaching disturbances. In order to avoid this, in the known method a control variable is derived from the extent of the temporarily stored data and supplied to the coder. This then reduces the quantization resolution of the image data blocks just coded, so that the data stream flowing into the buffer memory becomes smaller than that output. This prevents overflowing.
However, this proced
REFERENCES:
patent: 4581638 (1986-04-01), Chiariglione
patent: 4672441 (1987-06-01), Hoelzlwimmer
patent: 4704628 (1987-11-01), Chen
Amor Hamed
Luke Hans-Dieter
Britton Howard W.
Robert & Bosch GmbH
Striker Michael J.
LandOfFree
Method and circuit arrangement for reducing the data rate of dig does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and circuit arrangement for reducing the data rate of dig, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and circuit arrangement for reducing the data rate of dig will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-495578