| Abstrakt: |
Compression is a method of reducing data size by deleting redundancy of data and significant correlations between nearby pixels of images. Although the results of lossy compression reconstruction are not similar to the original image the compression technique is very important to save a large amount of memory and increase the speed of transmission, especially when dealing with images. In this work, a database of different five images was considered namely; Butterfly, car, Puppy, peppers, and house with sizes of (53, 47, 33, 44, and 51 Kb) respectively. The compression and classification of images were achieved by four different transform techniques Fast Fourier, Wavelet, Walsh, and Discrete cosine transform techniques with four compression ratios (10, 20, 40, and 80%) of the original image size. The results referred that the Fast Fourier transform FFT technique recorded a high compression images performance compared with other used techniques, which provided the highest average peak signal-noise ratio PSNR (30.55) and the lowest mean square error MSE (0.08) during 10% of compression ratio. [ABSTRACT FROM AUTHOR] |
