Objective: The purpose of this work is to present the experimental results by comparing the quality of different satellite images LANDSAT 7, MODIS and ASTER after compression, using four different compression methods. Methods/Analysis: The satellite images are compressed using four different basic compression methods namely, Pulse code Modulation, Differential Pulse Code Modulation, Discrete Cosine Transform and Sub band Coding. The Compression is performed with three different types of satellite sensor images having different spectral bands, picture bit-rate and level of details using VCDemo software package. Findings: The Mean Square Error, Signal to Noise Ratio and Peak Signal to Noise Ratio values are calculated to determine the quality of the images after compression. In order to find the quality of compression methods the Mean Square Error, Signal to Noise Ratio and Peak Signal to Noise Ratio values are collected for different satellite images by compressing with different bitrates. While comparing the obtained values for each compression methods, we found that the compression methods have different impacts in each satellite image according to the bit-rate used for compression and the level of details. Conclusion: The study proves that Discrete Cosine Transform and Sub band Coding’s performance are worthy for satellite image compression. Where Sub band Coding produces a very good Signal to Noise Ratio and Peak Signal to Noise Ratio values for all bit rates. Keywords: DCT, Image Compression, MSE, PSNR, Satellite Image Compression, SBC, SNR
Read full abstract