PENGEMBANGAN TEKNIK KOMPRESI HYBRID UNTUK DATA PENGINDERAAN JAUH MODIS

Abstract

Development on Remote Sensing field requires an optimal data compression technique to result high compression ratio but prevent data quality, so the data still be accepted on application remote sensing. One of the newest compression technique is hybrid compression that combine the reduction spectral redundancy and the reduction spatial redundancy. On this research will propose the improvement of hybrid technique compression for multiband MODIS data which developed by Yuan-Xiang Li, et.al (IEEE 2007). The improvement is by selecting better reference band for band prediction, skip the prediction band for unpredictable band and propose to use higher decomposition wavelet level will get better the compression performance. The Simulation result is the linier interband prediction with skip band will get the high value of PSNR ( 40 dB) and high compression ratio (80). The result of the compression will be used for remote sensing application of the MODIS data, specially for global earth observation. Key word: Compression, Hybrid, Prediction, Image,band, Wavelet, MODIS, PSNR, Symetry co-histogram