Abstract
Summary form only given. The contribution of this study is fourfold: (i) The author's own variable region vector quantization, which is a neurocomputation paradigm of the nearest neighbor type, is presented. (ii) By considering this neurocomputation paradigm and others, the total system is implemented as an emulator. The system is made up of a hypercube back end and its host. The host uses GHC for communication and control of the hypercube. Thus, this system uses an extended GHC, including commands for the fine-grained data parallelism. Such an extended version is called *GHC. The total system is tentatively called Neuro Cube, version 0. (iii) The author's algorithm is coded by *GHC and is executed for digital image compression on the above emulator. It is observed that the implemented two-level parallelism is quite effective for digital neurocomputation. (iv) The vector quantization is effectively combined with the backpropagation layered network to achieve efficient color image compression.
| Original language | English |
|---|---|
| Title of host publication | IJCNN Int Jt Conf Neural Network |
| Editors | Anon |
| Place of Publication | Piscataway, NJ, United States |
| Publisher | Publ by IEEE |
| Pages | 597 |
| Number of pages | 1 |
| Publication status | Published - 1989 |
| Externally published | Yes |
| Event | IJCNN International Joint Conference on Neural Networks - Washington, DC, USA Duration: 1989 Jun 18 → 1989 Jun 22 |
Other
| Other | IJCNN International Joint Conference on Neural Networks |
|---|---|
| City | Washington, DC, USA |
| Period | 89/6/18 → 89/6/22 |
ASJC Scopus subject areas
- Engineering(all)