| Presentation | 2009-12-14 Associative-Memory-Based LSI with Adaptive-Learning Capability Akio KAWABATA, Wataru IMAFUKU, Tania ANSARI, Hans Jurgen MATTAUSCH, Tetsushi KOIDE, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | When pattern recognition is achieved by conventional techniques, processing time becomes long and it is difficult to design an LSI. In the present research, the associative memory architecture for finding the most similar data among previously stored reference data is investigated for an application involving recognition and learning. We achieve high speed, low power consumption and a small area for the recognition function by using a mixed digital-analog fully parallel associative memory. For implementing the learning function of new reference data, we propose an associative memory based learning algorithm which imitates the concept of human's short/long-term memory. We apply the proposed learning algorithm to codebook-based image compression for evaluation and analysis of its efficiency. The created codebook with the proposed learning algorithm is evaluated for capturing the learning effect quantitatively with the Peak Signal Noise Ratio (PSNR). PSNR is an index of the image quality, and it can analyze the learning parameter dependence. In addition, we apply Huffman Coding to the codebook-based image compression, and verify that the compression ratio is improved from 12.8 to 14.1. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Associative Memory / Codebook-based Image Compression / Adaptive Learning / Huffman Coding |
| Paper # | ICD2009-93 |
| Date of Issue |
| Conference Information | |
| Committee | ICD |
|---|---|
| Conference Date | 2009/12/7(1days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | |
| Vice Chair | |
| Secretary | |
| Assistant | |
| Paper Information | |
| Registration To | Integrated Circuits and Devices (ICD) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Associative-Memory-Based LSI with Adaptive-Learning Capability |
| Sub Title (in English) | |
| Keyword(1) | Associative Memory |
| Keyword(2) | Codebook-based Image Compression |
| Keyword(3) | Adaptive Learning |
| Keyword(4) | Huffman Coding |
| 1st Author's Name | Akio KAWABATA |
| 1st Author's Affiliation | Research Institute for Nanodevice and Bio Systems, Hiroshima University() |
| 2nd Author's Name | Wataru IMAFUKU |
| 2nd Author's Affiliation | Research Institute for Nanodevice and Bio Systems, Hiroshima University |
| 3rd Author's Name | Tania ANSARI |
| 3rd Author's Affiliation | Research Institute for Nanodevice and Bio Systems, Hiroshima University |
| 4th Author's Name | Hans Jurgen MATTAUSCH |
| 4th Author's Affiliation | Research Institute for Nanodevice and Bio Systems, Hiroshima University |
| 5th Author's Name | Tetsushi KOIDE |
| 5th Author's Affiliation | Research Institute for Nanodevice and Bio Systems, Hiroshima University |
| Date | 2009-12-14 |
| Paper # | ICD2009-93 |
| Volume (vol) | vol.109 |
| Number (no) | 336 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |