Vol 3, No 2 (2000)
Basic sciences
Published online: 2003-05-27
Wavelet compression of nuclear medicine images
Nucl. Med. Rev 2000;3(2):153-156.
Abstract
BACKGROUND: The wavelet transform is a new mathematical
tool for the analysis of signals and images. The powerful lossy
compression techniques are built on the wavelets.
METHODS: The static liver scans (byte mode acquisition, 128 x 128 matrix size) were studied. Four level wavelet transform was performed for image compression by means of biorthogonal filters with linear phase characteristic. The SPIHT technique (Set Partition in Hierarchical Trees) was used to code the wavelet coefficients.
RESULTS: The compression ratios 90:1, with a good quality of decompressed images, were achieved (PSNR > 32).
CONCLUSIONS: The wavelet compression of scintigrams may be a useful practical tool because of substantial disk space saving and the reduction of transmission time. The decompression of images does not significantly affect their quality.
METHODS: The static liver scans (byte mode acquisition, 128 x 128 matrix size) were studied. Four level wavelet transform was performed for image compression by means of biorthogonal filters with linear phase characteristic. The SPIHT technique (Set Partition in Hierarchical Trees) was used to code the wavelet coefficients.
RESULTS: The compression ratios 90:1, with a good quality of decompressed images, were achieved (PSNR > 32).
CONCLUSIONS: The wavelet compression of scintigrams may be a useful practical tool because of substantial disk space saving and the reduction of transmission time. The decompression of images does not significantly affect their quality.
Keywords: wavelet transformimage procesingwavelet compression
