We present an optimization technique to select adaptively a set of control parameters for edge and boundary detection in intensity image data. The method is based on a genetic algorithm, and requires the definition of a performance measure to estimate the quality of the output in the absence of ground truth. Synthetic and real images are used to illustrate the approach.
|Number of pages||5|
|Journal||IEE Conference Publication|
|Issue number||443 pt 1|
|Publication status||Published - 1997|
|Event||Proceedings of the 1997 6th International Conference on Image Processing and its Applications. Part 1 (of 2) - Dublin, Irel|
Duration: 14 Jul 1997 → 17 Jul 1997