TWO DIMENSIONAL OPTIMUM EDGE RECOGNITION USING MATCHED AND WIENER FILTERS FOR MACHINE VISION.
In a previous paper, the authors examined the problem of optimum recognition of edges from a matched filter perspective and optimum one-dimensional edge filters were derived. It was shown that the common belief that good detection and good localization are in opposition is incorrect. Rather, there is a coupled relationship between detection and localization. In this paper, the one-dimensional results are extended to two dimensions. Optimum two-dimensional edge recognition consists of applying a one-dimensional matched filter in the edge normal direction and an orthogonal one-dimensional Wiener filter along the contour. The Wiener filter has an impulse response which is similar to that of commonly used smoothing filters. However, its purpose is not smoothing, but rather, to provide a best estimate of the edge contour.
|Title:||TWO DIMENSIONAL OPTIMUM EDGE RECOGNITION USING MATCHED AND WIENER FILTERS FOR MACHINE VISION.|
|UCL classification:||UCL > School of BEAMS > Faculty of Engineering Science
UCL > School of BEAMS > Faculty of Engineering Science > Computer Science
Archive Staff Only