This paper presents a lossy compression method for noisy images. The main contributions in this paper are : improve the performance of noisy image compression algorithms by using partial differential equations (PDEs) as an image preprocessing filter and show that it is more advantageous to apply a restoration filter before compressing noise images.
Method :The preprocessing filter is applied to the noisy image first. A comparing of the obtained outcomes with those in the scientific literature devoted to the restoration of noisy images is performed. The filtered image is then submitted to a compression algorithm. The compression algorithm employed is a hybrid of the following : DWT+SPIHT+HUFFMAN. 
Results:The simulation results demonstrate that the suggested technique is as efficient as approaches described in the scientific literature for picture compression and/or image restoration.