Boundary Detection via Dynamic Programming
Dana H. Ballard · Birkhäuser Basel eBooks · 1976
In this chapter we develop a technique for finding the boundaries of a candidate nodule. The technique is based on Montanari’s work (1970) on the extraction of curves with dynamic programming. We choose this approach because a) we can specify an intuitively satisfying mathematical objective function for the boundary which can be optimized, b) in a sense the solution is globally optimal, and c) the procedure is relatively insensitive to the noisy conditions encountered in the digitized radiograph. These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.