A performance measure of boundary detection algorithms
Philip L. Palmer, H.S. Dabis, Josef Kittler · 2002
In this paper we derive a quality measure for analysing the performance of algorithms which seek to find boundaries and boundary models between regions of differing mean grey-level value. The measure can be used to compare different approaches and, more importantly, it can also be used as a first step to building self-optimising vision systems that can automatically optimise important parameters at each level of the system.