Abstract

Image compression aims to reduce the storage and transmission requirements of digital images. This work is implemented by block-based image compression which used 2D Discrete Cosine Transform and this technique is widely used in JPEG. The input is grayscale image. The grayscale image is divided into 8x8 blocks. Each block is converted from spatial domain to frequency domain using Discrete Cosine Transform. some of the high frequency information which are less significant are removed from the image and low frequency information which are more significant are kept to make perfect compression of the given signal. Then Inverse Discrete Cosine Transform is applied to the low frequency data to rebuild the image from compressed frequency data. The quality of compressed image is computed using Peak Signal-to-Noise Ratio (PSNR). The result shows that better compression can be made with minimal loss in optical quality and high spot the success of 2D Discrete Cosine Transform in applications of image compression.