In  Noble uses mathematical morphology to find image structure. Several different morphological operations are described; these are used to enhance edges and find two dimensional features. The ``erode-dilate'' operator is similar to a first order derivative, and the ``open-close'' operator is similar to a second order derivative. Good quality edges are found by tracking both sides of each edge and then stitching these ``half boundaries'' together. Connectivity at junctions is good, although spurious short ``tails'' sometimes appear at structures such as ``T'' junctions. The algorithm, including edge tracking, is fairly computationally expensive.