Using　the　significant　spectral　correlation　within　the　hyperspectral　images,　we　present　a　lossless　compression　algorithm　in　this　paper.　The　method　includes　four　key　steps:　(1)　band　ordering　according　to　spectral　correlation　coefficient;　(2)　2-D　and　3-D　hybrid　adaptive　prediction　based　on　local　texture　and　neural　networks,　respectively;　3)　hybrid　prediction　based　on　the　correlation　of　neighborhood;　4)　context-based　Golomb　coding.　Experimental　results　show　that　this　method　can　remove　the　spatial　and　spectral　redundancy　efficiently　and　outperforms　JPEG-LS　and　3D-APA　on　average　bit　rate　obviously.