Pulse or digital communications – Bandwidth reduction or expansion – Television or motion video signal
Reexamination Certificate
1998-09-28
2002-03-19
Britton, Howard (Department: 2613)
Pulse or digital communications
Bandwidth reduction or expansion
Television or motion video signal
C375S240110, C375S240190
Reexamination Certificate
active
06359928
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates generally to a system and method for compressing images, and specifically to a system and method for performing image compression using multi-threshold wavelet coding.
2. Description of Related Art
The compression of images for transmission or storage has become widely practiced in a variety of contexts in order to reduce the rather large size of most image files. Progressive fully-embedded image compression has become particularly important when images are being compressed for real-time transmission, especially in multimedia environments for video conferencing, video games, Internet image transmissions, digital TV and the like. There are two types of compression methods, lossless and lossy techniques. A lossless technique allows exact reconstruction of all of the individual pixel values, while a lossy technique does not allow for an exact reconstruction of the image. Much higher compression ratios can be achieved by lossy techniques if some loss of the exact pixel values can be tolerated.
Image compression is accomplished using an image coder, wherein a typical transform image coder consists of three parts: a transform
10
, a quantizer
12
, and an encoder
14
, as illustrated in FIG.
1
. The transform
10
decorrelates the original image from the spatial domain into a transform domain where the image can be more compactly represented. The quantizer
12
restricts the representation of the signal into discrete intervals, in which sampled image pixel data values, like color and luminance, are represented by, or are mapped onto, a fixed number of predefined quantizer values. The quantized signal is composed of quantizer values that are, in fact, approximations of the sampled image pixel values. Therefore, the encoding of the image data onto a limited number of quantizer values necessarily produces some loss in accuracy and distortion of the signal after the decoding process. Finally, the image coder encodes the quantized coefficients using the encoder
14
. One of the most widely used transform image coding schemes is the Joint Photographers Expert Group (JPEG) standard formed in collaborative effort of the CCITT (International Telegraph and Telephone Consultative Committee) and the ISO (International Standards Organization).
JPEG utilizes a block discrete cosine transform (DCT) as its transform component. Block DCT and other similar transforms decompose images into a representation in which each coefficient corresponds to a fixed size spatial area and a fixed frequency bandwidth, where the bandwidth and spatial area are effectively the same for all coefficients in the representation. Edge or object boundary information in an image tends to disperse so that many non-zero coefficients are required to represent edges with good precision. Since the edges represent relatively insignificant energy with respect to the entire image, transform coders using block DCT are fairly successful at high bit rates. However, at low bit rates, transform coders using block DCT, such as JPEG, tend to allocate too many bits to areas of high statistical spatial correlation and have few bits left over to represent the edges. As a result, blocking artifacts often result with transform coders using block DCT, which significantly degrades the resolution of the reconstructed image. It is well known that, due to the use of block DCT, JPEG tends to introduce severe blocking artifacts at low bit rates
Recently, several embedded wavelet coders have been proposed which use a global wavelet transform instead of block DCT. A wavelet transform is a type of transform which transfers image pixel values from the spatial domain to the frequency domain to perform the decorrelation. Wavelet transforms divide an original image into a plurality of frequency bands (subbands) using various types of filtering devices. Each of the subbands contains a different scale of wavelet coefficients, wherein the number of subbands created depends upon the desired scale to be reached. The first advantage of embedded wavelet coders is that they give a much better rate-distortion performance than JPEG and the blocking artifact is reduced significantly. The second advantage of embedded wavelet coders is their progressive coding property due to the use of successive approximation quantization (SAQ) and bit layer coding. SAQ is used to determine which quantization bits are significant and which are insignificant. Since there are correlations between insignificant bits, several methods have been developed to predict or to group these insignificant bits for effective lossless entropy coding.
During the quantization step in most of the proposed embedded wavelet coding schemes, a single initial quantization threshold is applied uniformly to wavelet coefficients of all subbands in the coding schemes to locate the significant bits. In most images, the maximum magnitude of the wavelet coefficients varies significantly from subband to subband. Coefficients in the lower frequency subbands tend to have a higher magnitude, while coefficients in higher frequency subbands have a smaller magnitude. With a single initial quantization threshold, it is common to generate a lot of zeros in the high frequency subbands during the first quantization steps. Thus, current quantization schemes generate too many insignificant bits in the first several quantization steps which unnecessarily adds to the computational complexity and coding time of the coding process.
There is clearly a need for a system and method for performing image compression using multi-threshold wavelet coding which improves the quality of a reconstructed image by eliminating the blocking artifact frequently occurring at low bit rates. Moreover, there is a need for a system and method for performing image compression using multi-threshold wavelet coding which reduces the computational complexity of the coder and improves the efficiency of the coder.
SUMMARY OF THE INVENTION
It is a primary object of the present invention to overcome the aforementioned shortcomings associated with the prior art.
The present invention provides a system and method for performing image compression using multi-threshold wavelet coding which avoids blocking artifacts from being produced in a reconstruction of the compressed image.
The present invention further provides a system and method for performing image compression using multi-threshold wavelet coding having a reduced computational complexity and an improved efficiency over previous image coders.
The system and method for performing image compression using multi-threshold wavelet coding of the present invention provides a better rate-distortion performance at high-bit rates while still reducing computational complexity.
The present invention provides a system and method for performing image compression using multi-threshold wavelet coding (MTWC) which utilizes a separate initial quantization threshold for each subband generated by the wavelet transform, which substantially reduces the number of insignificant bits generated during the initial quantization steps. Further, the MTWC system chooses the order in which the subbands are encoded according to a preferred rate-distortion tradeoff in order to enhance the image fidelity. Moreover, the MTWC system utilizes a novel quantization sequence order in order to optimize the amount of error energy reduction in significant and refinement maps generated during the quantization step. Thus, the MTWC system provides a better bit rate-distortion tradeoff and performs faster than existing state-of-the-art wavelet coders.
REFERENCES:
patent: 5101446 (1992-03-01), Resnikoff et al.
patent: 5128757 (1992-07-01), Citta et al.
patent: 5429636 (1995-07-01), Shikhman et al.
patent: 5552832 (1996-09-01), Astle
patent: 5555511 (1996-09-01), Ooi
patent: 5657085 (1997-08-01), Katto
patent: 5798794 (1998-08-01), Takahashi
patent: 5953460 (1999-09-01), Wu
patent: 6160846 (2000-12-01), Chiang et al.
patent: 490539 (1992-06-01), None
patent: 490539 (199
Bao Yiliang
Chen Homer H.
Kuo Chung-Chieh
Wang Houng-Jyh Mike
Akin Gump Strauss Hauer & Feld & LLP
Britton Howard
Rourk Christopher J.
University of Southern California
LandOfFree
System and method for compressing images using... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System and method for compressing images using..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System and method for compressing images using... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2867573