Compression of line drawing images using Hough transform for exploiting global dependencies

A two-stage method is proposed for compressing bi-level line drawing images. In the first stage line elements are extracted from the image using the Hough transform. In the second stage the original image is compressed by a pixelwise context model. The compressed file consists of the extracted line elements and the compressed raster image. Pixels in the reconstructed feature image are utilized in the local context model in the form of improved prediction accuracy. Better compression performance is achieved if the improvement outweighs the overhead required by the line elements.