Image analysis – Image compression or coding – Gray level to binary coding
Reexamination Certificate
1998-01-09
2001-03-27
Mancuso, Joseph (Department: 2623)
Image analysis
Image compression or coding
Gray level to binary coding
C382S232000
Reexamination Certificate
active
06208761
ABSTRACT:
TECHNICAL FIELD
The present invention is related to a methods and devices for compressing and coding digitally represented moving pictures, grey scale as well as colour, which are to be transmitted on a channel and in particular a channel which has a relatively small capacity or bandwidth as well as a system for transferring moving pictures using compression of the pictures.
BACKGROUND AND PRIOR ART
In many applications involving transmission of video signals, the capacity of the channel used is limited. A digitalized video image, however, consists of a very large number of bits. Hence, when transmitting such an image, consisting of a very large number of bits, over a channel which has a limited bandwidth on which data traffic having a low or a very low bit rate can be transmitted, transmission times for most applications become unacceptably long, if every bit of the image has to be transmitted. This is especially true i n the case of moving pictures, where strict real time constraints exist.
Therefore, much research efforts in recent years have concerned coding methods and techniques for digitalized images, aiming at reducing the number of bits to be transmitted as much as possible.
These methods can be divided into two groups:
Lossless methods, i.e. methods exploiting the redundancy in the image in such a manner that the image can be reconstructed by the receiver without any loss of information, i.e. the reconstructed image coincides exactly with the original image.
Lossy methods, i.e. methods exploiting the fact that all bits are not equally important to the receiver. In these methods the received image is not identical to the original, but looks, e.g. for the human eye, sufficiently alike the original image.
In the field of video coding for low and very low bit rate applications, many of the present approaches involve the use of the frequency domain, e.g. by means of the Discrete Cosine Transform (DCT).
These approaches led to different video coding standards: JPEG (Joint Photographic Experts Group) for still image compression; MPEG I/II (Moving Picture Experts Group) for storing and multimedia applications and for video coding at medium—high bit rates; CCITT Recommendation H.261 (Px64) for real time video coding of video conferences, remote surveillance, and video telephone applications; CCITT Recommendation H.263, which is related to H.261, but is developed for lower bit rates (16-64 kilo bits per second kbps).
Other techniques are, for example, based on wavelets and subband coding methods, fractal transformations, vector quantization, etc. These compression methods work quite well in many conditions. However, these methods have problems at very high compression ratios, such as in the area of real-time transmission of moving pictures over channels having low capacities, such as the public telephone lines.
Although the efforts and improvements made, the compression of the images in relation to the achieved quality of the resulting image, the compression/quality ratio, still remains quite low and not good enough for many applications, especially in the case of narrow band channels having capacities less then 64 kbps, such as the public telephone lines. In particular problems remain in the introduction of low-pass effects, blocking effects and undesired frequencies.
Also, the existing coding methods are based on computationally complex and expensive systems, comprising frequency or fractal transformations, filtering stages and vector quantization processes. In order to perform these algorithms in real time expensive processors such as DCT processors, zigzag processors, blocking processors etc. are required.
Another possible approach is the bit plane coding technique.
This method maps the pixels of a digitalized image into a number of binary bit planes the first one of which usually consists of the most significant bits of the pixels. Thus, the image which consists of pixels, which in turn consist of a number of bits, are mapped into a number of bit planes, where the number of bit planes is equal to the number of bits per pixel (bpp).
The purpose of mapping the bits into bit planes is to exploit the spatial redundancy of the digitalized video image. When exploiting, by means of compression, these redundancies no information is lost, and thus the images compressed and transmitted using this technique can be recreated exactly bit by bit, i.e. the technique is lossless. This kind of techniques has been introduced and successfully applied in cases of lossless coding of still pictures, such as X-ray medical images, satellite and space images and facsimile images.
The method of bit plane coding is quite efficient compared to other lossless existing coding methods for coding still images.
In the book “Digital Image Compression Techniques”, M. Rabbani, P. W. Jones, SPIE Optical Engineering Press Washington, 1991, the bit plane coding technique is described. The book also deals with Gray code in the field of bit plane coding, for lossless coding of still images.
Further, in U.S. Pat. No. 5,374,945 a method is disclosed for grey level printing, in which several bit plane datasets comprise a bogus bit plane, most significant bit plane, next-to-most significant bit plane, least significant bit plane, and next-to-least significant bit plane and a significance of each said bit plane corresponds to a printing time length.
U.S. Pat. No. 5,142,619 patent discloses techniques using the XOR-operation. A device described in the document has means provided for comparing the contents of two bit planes in order to compare the respective pixel locations and exclusive OR-ing each pair of corresponding pixels to set a corresponding pixel location in a third bit plane to reflect similarity or dissimilarity between the compared pair of pixels of two bit planes of the same image.
U.S. Pat. No. 4,653,112 concerns image data management, where image data are organized in bit planes. Data comprising the most to the least significant bits are arranged in the first to the last bit planes respectively.
U.S. Pat. No. 4,546,385 relates to data compression of graphic images. A graphic image has at least first (most significant bit) and second (least significant bit) bit planes. The most significant bit of a pixel and the successive pixel are compared using an exclusive OR-operation on a spatial dimension, i.e. the XOR operation is performed between pixels of the same image.
The European patent application EP-A1 0 547 528 discloses coding of binary bit planes eliminating the need for forming a Gray code bit plane representation. The invention uses different significances for different bit planes.
SUMMARY
It is an object of the present invention to overcome the disadvantages and problems associated with the methods described above, and to provide methods and devices by means of which moving pictures of a good quality can be compressed for transmission and storage purposes.
This object is achieved by the present invention the characteristics of which are set out in the appended claims.
Thus, a method is provided for coding both grey scale and colour images, in particular moving pictures. This method is also based on bit plane coding, and can transmit moving pictures of relatively good quality only requiring bit rates of approximately 10 kbps. The method for achieving this can be divided into the following five substeps:
I. Representing the digitalized images by means of the bit plane representation, and skipping from the achieved result the planes corresponding to the bits that carry the least information, i.e. the least significant planes (plane skipping).
II. Coding the remaining bit planes with the Gray code, and when doing this, in order to maximally exploit the redundancy of the bit planes and to reduce the dynamic of the resulting signal, before making use of the Gray code, shifting the bits of the remaining planes into the lowest positions and then padding the vacant positions with zeroes.
III. Reduction the number of bits needed to be transmitted by means of segmentation of the frames and by applying a moti
Passaggio Filippo
Suanno Guiseppe
Bali Vikkram
Mancuso Joseph
Nixon & Vanderhye P.C.
Telefonaktiebolaget LM Ericsson (publ)
LandOfFree
Video coding does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Video coding, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Video coding will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2487882