Image compression based on finite-state vector quantization
Thesis/Dissertation
·
OSTI ID:5145770
Compression of imagery is becoming increasingly important in digital transmission and storage. Recently, vector quantization (VQ) has emerged as a popular tool in image-compression algorithms. VQ is a powerful pattern-matching technique which jointly encodes blocks (or vectors) of input samples or features, using a codebook of prototype patterns. A technique of enchancing the power of VQ is to use information obtained from previously encoded vectors in the encoding of each successive input vector. In this work, the author studies new algorithms based on FSVQ for monochrome image compression at rates below 0.5 bpp. He introduces a novel formulation of the state and state-transition rule which uses a perceptually based classifier. Two types of finite-state VQ coders are studied: fixed-rate coders where each input block is encoded with the same number of bits, and variable-rate coders where the number of bits used to encode each block is varied according to the detail in the input image.
- Research Organization:
- California Univ., Santa Barbara, CA (USA)
- OSTI ID:
- 5145770
- Country of Publication:
- United States
- Language:
- English
Similar Records
Low-rate image coding using vector quantization
Implementation of VQ algorithms on a reconfigurable array processor. Professional paper
Implementation of VG algorithms on a reconfigurable array processor
Thesis/Dissertation
·
Sun Dec 31 23:00:00 EST 1989
·
OSTI ID:6155745
Implementation of VQ algorithms on a reconfigurable array processor. Professional paper
Technical Report
·
Wed May 01 00:00:00 EDT 1991
·
OSTI ID:5217551
Implementation of VG algorithms on a reconfigurable array processor
Technical Report
·
Fri Nov 30 23:00:00 EST 1990
·
OSTI ID:5673441