Autonomous Band Registration for On-Board Applications

Yu, Guoxia; Vladimirova, Tanya; Sweeting, Martin

doi:10.1109/icspc.2007.4728572

Cited by 7 publications

(2 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The bands of multispectral images are not aligned accurately. So image registration [8] is required before compression with 3-D prediction. Before the compression of the current band, the previous band is already compressed and taken as the reference band, while the first band is compressed using the intra-band mode.…”

Section: Evaluation Of the Compression Performancementioning

confidence: 99%

An Efficient On-Board Lossless Compression Design for Remote Sensing Image Data

Vladimirova

Sweeting

2008

IGARSS 2008 - 2008 IEEE International Geoscience and Remote Sensing Symposium

Self Cite

View full text Add to dashboard Cite

In the Consultative Committee for Space Data SystemsLossless Data Compression (CCSDS-LDC) algorithm a small region of image data is compressed independently to achieve error-resilience functionality, which is one of the most distinct requirements for space applications. 2-D or 3-D predictions are believed to bring superior compression performance, however they interfere with the error-resilience property. This paper proposes a new scan scheme to address the problem. By combination of multidimensional predication and embedded radiometric calibration technique, the proposed system performs very effectively on remote sensing panchromatic and multispectral images. Finally a high-level reconfigurable model for the hardware implementation of the proposed system is developed targeted at FPGA devices.

show abstract

Section: Evaluation Of the Compression Performancementioning

confidence: 99%

An Efficient On-Board Lossless Compression Design for Remote Sensing Image Data

Vladimirova

Sweeting

2008

IGARSS 2008 - 2008 IEEE International Geoscience and Remote Sensing Symposium

Self Cite

View full text Add to dashboard Cite

show abstract

“…6 out of 8 available bands are used, as they are sharing the same resolution of 30 m. The bands of multispectral images are not aligned accurately. So image registration [7] is required before compression with 3-D prediction. Before the compression of the current band, the previous band is already compressed and taken as the reference band, while the first band is compressed using the intra-band mode.…”

Section: Evaluation Of Compression Performancementioning

confidence: 99%