Compression For 2-D Spectra Images Based On LAMOST

LAMOST, which is denoted to"Large Sky Area Multi-Object Fiber Spectroscopy Telescope", is one of the National Major Scientific Projects undertaken by the Chinese Academy of Science.LAMOST will obtain several ten-thousands of spectra per night and the data volume will be several gigabytes. Therefore, compression of such large amounts of data is an important consideration for large scale digital sky surveys. Higher compression ratio not only can reduce the volume of data needed to store, but also can shorten the time required to transmit the data. This paper introduces several lossless compression methods for spectra images. The main works are described as follows:1. We study the characteristics of 2-D spectra images according to observational model and data-processing model. The conditionality of compressibility is dicussed from a view of information entropy.2. Based on the structural characteristics of 2-D spectra images, an improved Hcompress algorithm is proposed for lossless compression. Performance improvement can be achieved by making use spectra characteristic in the wavelength direction.3. Applied IWT (Integer Wavelet Transform) to lossless compression for 2-D spectra images. EZW is employed to describe the zero-tree structure of wavelet coefficients, the resulting stream put through EZW encoder can be transformed to character string including only five characters that is easily compressed by entropy coding.4. Applied the JPEG2000 to multi-exposure spectra images. Each 16-bits spectra image is divided into two images, the lower 8-bits forms one image and the higher 8-bits forms another. So three exposures of LAMOST can be divided into two RGB images. Some improvements can be achieved by removing the correlation between different exposures that characterize the same object.