LOSSY COMPRESSION OF HYPERSPECTRAL IMAGES USING ONLINE LEARNING BASED SPARSE CODING
International Workshop on Computational Intelligence for Multimedia Understanding (IWCIM), Paris, Fransa, 1 - 02 Kasım 2014, (Tam Metin Bildiri)
- Yayın Türü: Bildiri / Tam Metin Bildiri
- Doi Numarası: 10.1109/iwcim.2014.7008809
- Basıldığı Şehir: Paris
- Basıldığı Ülke: Fransa
- Anahtar Kelimeler: Sparse Coding, Hyperspectral Imagery, Anomaly Detection, Online Learning, CLASSIFICATION
- Ankara Üniversitesi Adresli: Hayır
Özet
A lossy hyperspectral image compression method is proposed using online learning based sparse coding. The least number of coefficients are obtained to represent hyperspectral images by applying the sparse coding algorithm which is based on a dicriminative online dictionary learning method. Results indicate that a pre-analysis of the number of non-zero dictionary elements may help in improving the overall compression quality.