Adaptive compression of DICOM-image data

In this article a method to classify digital images into three categories based on a provisional analysis of the image content and a subsequent compression with the help of a suitable algorithm of compression is proposed. To classify the images two parameters are used. The first parameter carries the frequency information about the image. It represents the mean of the absolute amplitude of the Wavelet coefficients in the high frequency parts of the spectrum. The second parameter is an indicator concerning information about the structure of the image. The second parameter is constituted through the entropy of the length of the segments in one line and the entropy of the length of the segments in one column. In this article the results for the check of those classification rules for DICOM images are given as a confirmation of the effectivness of the method proposed. The implementation of the image classification algorithm an the compression algorithms in the modelling process is performed in JAVA