Shock filters are based in the idea to apply locally either a dilation or an erosion process, depending on whether the pixel belongs to the influence zone of a maximum or a minimum. They create a sharp shock between two influence zones and produce piecewise constant segmentations. In this paper we design specific shock filters for the enhancement of coherent flow-like structures. They are based on the idea to combine shock filtering with the robust orientation estimation by means of the structure tensor. Experiments with greyscale and colour images show that these novel filters may outperform previous shock filters as well as coherence-enhancing diffusion filters.