A performance evaluation on DCT and wavelet-based compression methods for remote sensing images based on image content

Abstract

This study compared the performance of DCT (Discrete Cosine Transform) and wavelet-based compression methods over remote sensing images. For the experiment, we collected two sets of test images, consisting of 20 true color remote sensing images each, to explore how DCT and wavelet-based methods perform over the different types of images: natural scenes and urban scenes. Each test image was encoded and decoded with the JPEG and JPEG 2000 codecs at the compression ratios of 2∶1, 5∶1, 10∶1, and 20∶1, and the results were analyzed with the amount of errors introduced from the encoring process. According to the results, the wavelet-based method worked better than the DCT method, and both of the methods appeared to work slightly better for the images of urban scene.