Gray image compression based on quad tree and partial differential equation

A novel gray image compression algorithm based on quad tree and partial differential equation (PDE) is proposed, which only sends and processes some scattered data rather than the entire image. On the encoder, the quad tree is adopted to divide an image into sub-images, and some pixels are selected to encode. On the decoder, a PDE-based image interpolation method is applied to reconstruct the image, which is effective to reduce artifact. Experimental results show that the proposed method provides better performance in compression ratios and PSNR, especially for larger and non-texture images. And it can also keep details where pixel gray value varies hugely and remove block-effect conveniently. Therefore it has high practical values.