J. Levman, J. Alirezaie, and G. Khan (Canada)
Image processing, edge detection, multi-species genetic algorithm
In this paper, we present a novel technique for image edge detection. Most computer vision systems rely on the effectiveness of an image processing application's edge detection abilities. In addition, edge detection is used in image enhancement. Significant research has been performed on various approaches to edge detection. A fair number of approaches utilize a genetic algorithm in the edge detection process. Our technique is the only approach we are aware of that incorporates a symbiotic dual-species genetic algorithm. This dual-species genetic approach attempts to optimize the edge detection mask from classical methods. Our method requires only a simple set of input data to be used for training the genetic algorithm. The results show a marked improvement over a similar genetic based edge detection technique and over more traditional methods. It is also noted that the above approach is not restricted simply to edge detection, but can also be used for generic spatial filtering.
Important Links:
Go Back
