Frequency adaptive learning SOFM for grayscale image compression
A new learning algorithm is proposed where the learning constant is based on the localized winning frequency of each individual neuron. Blocks of sub-image are fed in a butterfly-jumping sequence during the training process. Untrained neurons are also re-initialized.
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Published: |
2008
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10356/3336 |