Reversible Compression of Multispectral Imagery Based on an Enhanced Inter-Band JPEG Prediction
Conference Paper
Publication Date:
1997
abstract:
A modified inter-band scheme based on the predictors used by lossless JPEG is proposed for lossless data compression of multispectral images. Basically, the value of the current pixel in the current band is predicted by the best JPEG predictor on the previously encoded band. Improvements are achieved by considering also the corresponding prediction error on the previous band. Coding performances, assessed on a variety of Landsat TM data, are better by about 3% than those of the basic inter-band scheme, and by 7% than those attained by lossless JPEG.
Iris type:
04.01 Contributo in Atti di convegno
Keywords:
Reversible image compression; multispectral imagery; inter-band prediction; enhanced lossless JPEG; Landsat TM
List of contributors:
Alparone, Luciano; Aiazzi, Bruno; Baronti, Stefano
Book title:
Proceedings of IEEE IGARSS '97: Remote Sensing - A Scientific Vision for Sustainable Development