Parametric Feature Detection

Nayar, Shree K.; Baker, Simon; Murase, Hiroshi

We propose an algorithm to automatically construct feature detectors for arbitrary parametric features. To obtain a high level of robustness we advocate the use of realistic multi-parameter feature models and incorporate optical and sensing effects. Each feature is represented as a densely sampled parametric manifold in a low dimensional subspace of a Hilbert space. During detection, the brightness distribution around each image pixel is projected into the subspace. If the projection lies sufficiently close to the feature manifold, the feature is detected and the location of the closest manifold point yields the feature parameters. The concepts of parameter reduction by normalization, dimension reduction, pattern rejection, and heuristic search are all employed to achieve the required efficiency. By applying the algorithm to appropriate parametric feature models, detectors have been constructed for five features, namely, step edge, roof edge, line, corner, and circular disc. Detailed experiments are reported on the robustness of detection and the accuracy of parameter estimation. In the case of the step edge, our results are compared with those obtained using popular detectors. We conclude with a brief discussion on the use of relaxation to rene outputs from multiple feature detectors, and sketch a hardware architecture for a general feature detection machine.



More About This Work

Academic Units
Computer Science
Department of Computer Science, Columbia University
Columbia University Computer Science Technical Reports, CUCS-028-95
Published Here
February 10, 2012