In this work a new non-linear transform to analyze the non-linearities in an image has been proposed. In an image there are both high and low frequency components. The low frequency components in an image can be easily compressed than that of the high frequency components. So a non-linear transform called Peak Transform has been designed which minimizes the high frequency components in the image to a greater extent, thus making the image to get compressed to a greater extent. The Peak Transform along with Wavelet Transform and sub-band decomposition, it is possible to reduce the signal energy in high frequency sub-bands to get the significant transform coding gain. This can be used to efficiently represent and compress data or images. The Peak Transform image encoder outperforms state-of-the-art JPEG2000 and H.264 (INTRA) encoders by up to 2-3 dB in peak signal-tonoise ratio (PSNR), especially for images with a significant amount of high-frequency components. The proposed nonlinear transform called Peak Transform to convert hard-to-compress signals into easy-to-compress ones will be very useful in efficiently representing the data.
展开▼