<title>Comparison of the effect of lossy compressions with the modulation transfer function</title>

Abstract

The purpose of this study was to examine the effects of lossy image compression on the quality of digital radiographic images and to compare the resolution properties of the three image compression algorithms (JPEG, wavelet and fractal) on digital radiographic images using the Modulation Transfer Function(MTF). A single point source of Tc 99m was scanned by a Gamma camera and its image was compressed by JPEG, wavelet and fractal algorithms at various compression levels. The Point Spread Functions (PSFs) for the original and compressed images were generated in Matlab 5.0. This was done by analyzing the pixel values along a direction through the center of the point source on each image. The MTF for each image was then calculated by Fourier transform of its Point Spread Function (PSF). The resolution properties of each compression techniques were presented. It has been found that the resolution properties of the compressed images depend on both the compression levels and algorithms. The correlation of image quality, compression levels and algorithms are not straightforward.