2005 Fiscal Year Final Research Report Summary
Constructing a general image codebook using vector quantization and application to mobile communication terminals.
| Project/Area Number |
16500095
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Perception information processing/Intelligent robotics
|
|---|
| Research Institution | Muroran Institute of Technology |
|---|
Principal Investigator |
SUZUKI Yukinori Muroran Institue of Technology, Professor, 工学部, 教授 (00179269)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
MAEDA Junji Muroran Institue of Technology, Professor, 工学部, 教授 (00002311)
SAGA Sato Muroran Institue of Technology, Associate Professor, 工学部, 助教授 (90270793)
WATANABE Osamu Muroran Institue of Technology, Assistant Professor, 工学部, 講師 (50343017)
|
|---|
| Project Period (FY) |
2004 – 2005
|
| Keywords | vector quantization / image compression / mobile communication / frantal dimension |
|---|
| Research Abstract |
Telecommunication networks are spreading worldwide. People can communicate with each other beyond spatial restrictions using the Internet. In this situation, the demand for transmission bandwidth and storage space continues to outstrip the capacity of existing technologies. Image and video compression technology is therefore essential for the effective use of communication networks. In this research project, we have developed a new method to compress images using vector quantization. The sizes of the training vectors to prepare a codebook are determined on the basis of local fractal dimensions (LFDs). In this sense, each block size to divide the learning image is determined by the LFDs. In principle, the smaller the number of pixels in a training vector is, the larger is a codebook (low bit rate). Furthermore, the smaller the number of pixels in a codebook is, the higher is the quality of the encoded image. This is a tradeoff between compression rate and quality of the encoded image. We assume that this tradeoff can be solved by a method of variable block sizes. The block size to divide an area of an image is carried out using a method of discriminant analysis. The range of LFD values is 2.0 to 3.0. We divide the histogram of LFDs into F clusters. We then prepared sub-codebooks each of which corresponds to individual code-vector size. By this division, we construct a codebook to encode images.
Furthermore, the code-vectors are computed from the training vectors using a generalized Lloyd algorithm (GLA). The performance of the proposed algorithm was evaluated by compression rate and quality of encoded images in comparison with those using other algorithms to construct a codebook. The experiments showed that the proposed algorithm solved the tradeoff between compression rate and quality of image. The newly developed algorithm could be useful for image communication for mobile communication terminals.
|
