Lossless compression using joint predictor for astronomical images

研究成果: 書貢獻/報告類型會議論文篇章同行評審


Downloading astronomical images through Internet is a slow operation due to their huge size. Although several lossless image coding standards that have good performance have been developed in the past years, none of them are specifically designed for astronomical data. Motivated by this, this paper proposes a lossless coding scheme for astronomical image compressions. We design a joint predictor which combines the interpolation predictor and partial MMSE predictor. Such strategy benefits from its high compression ratio and low computation complexity. Moreover, the scalable and embedding functions can be further supported. The interpolation predictor is realized by upsampling the downsampled input image using bi-cubic interpolation, while the partial minimum mean square error (MMSE) predictor predicts the background and foreground (i.e., stars) separately. Finally, we design a simplified Tier-1 coder from JPEG2000 for entropy coding. Our experimental results show that the proposed encoder can achieve higher compression ratio than JPEG2000 and JPEG-LS.

主出版物標題Advances in Visual Computing - 5th International Symposium, ISVC 2009, Proceedings
版本PART 2
出版狀態已出版 - 2009
事件5th International Symposium on Advances in Visual Computing, ISVC 2009 - Las Vegas, NV, United States
持續時間: 30 11月 20092 12月 2009


名字Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
號碼PART 2
5876 LNCS


???event.eventtypes.event.conference???5th International Symposium on Advances in Visual Computing, ISVC 2009
國家/地區United States
城市Las Vegas, NV


深入研究「Lossless compression using joint predictor for astronomical images」主題。共同形成了獨特的指紋。