Most compression methods have been developed to minimize mean squared errors. On the other hand, discriminant information required to distinguish between various classes is also vital for applications where hyperspectral images are used for classification. However, the discriminant information may not be well preserved during lossy compression process since it may not be large in energy. To deal with this problem, a hybrid compression method was proposed. In this paper, we propose a method to optimize signal and discriminant information for the hybrid method. Experimental results show some improvement can be achieved by this optimization.
|Title of host publication||IISA 2016 - 7th International Conference on Information, Intelligence, Systems and Applications|
|Publisher||Institute of Electrical and Electronics Engineers Inc.|
|Publication status||Published - 2016 Dec 14|
|Event||7th International Conference on Information, Intelligence, Systems and Applications, IISA 2016 - Chalkidiki, Greece|
Duration: 2016 Jul 13 → 2016 Jul 15
|Name||IISA 2016 - 7th International Conference on Information, Intelligence, Systems and Applications|
|Other||7th International Conference on Information, Intelligence, Systems and Applications, IISA 2016|
|Period||16/7/13 → 16/7/15|
Bibliographical notePublisher Copyright:
© 2016 IEEE.
All Science Journal Classification (ASJC) codes
- Information Systems
- Computer Science Applications
- Artificial Intelligence
- Social Sciences (miscellaneous)