The present invention relates to the image compression and restoring method for binary images that can be used in binary image equipment including printers.In the present invention, an image is not compressed by each pixel but by cell that combines four pixels. And each cell is compressed in two steps. In the first step, the number of black pixels contained in the cell is compressed. And in the second step, using the number of pixels the information compressed in the previous step, the exact location of pixels that constitute the cell is compressed. By doing this, the compression rate can be increased. And even when the compressed bit rows are restored in the first step, image restoration is available with just a small degradation.