Unsupervised edge map scoring: a statistical complexity approach
This addresses the need for unsupervised edge map evaluation in computer vision, offering a method for parameter tuning and algorithm comparison, though it is incremental as it builds on existing statistical approaches.
The authors tackled the problem of evaluating edge maps without ground truth by proposing a new Statistical Complexity Measure (SCM) based on equilibrium and entropy indices, which outperformed Pratt's Figure of Merit on South Florida and Berkeley databases.
We propose a new Statistical Complexity Measure (SCM) to qualify edge maps without Ground Truth (GT) knowledge. The measure is the product of two indices, an \emph{Equilibrium} index $\mathcal{E}$ obtained by projecting the edge map into a family of edge patterns, and an \emph{Entropy} index $\mathcal{H}$, defined as a function of the Kolmogorov Smirnov (KS) statistic. This new measure can be used for performance characterization which includes: (i)~the specific evaluation of an algorithm (intra-technique process) in order to identify its best parameters, and (ii)~the comparison of different algorithms (inter-technique process) in order to classify them according to their quality. Results made over images of the South Florida and Berkeley databases show that our approach significantly improves over Pratt's Figure of Merit (PFoM) which is the objective reference-based edge map evaluation standard, as it takes into account more features in its evaluation.