직선의 불변 기하위상 전파를 통한 물체인식 = Fast Object Recognition using Local Energy Propagation from Combination of Salient Line Groups

다국어 초록 (Multilingual Abstract)

We propose a dynamic programming-based formulation for recognizing and matching line patterns by defining a robust and stable geometric representation that is based on the perceptual organizations. Usually, the endpoint proximity and co-linearity of image lines, as two main perceptual organization groups, are useful cues to match the model shape in the scene. As the endpoint proximity, we detect junctions from image lines. We then search for junction groups by using geometric constraint between the junctions, which has a geometric invariant property. A junction chain similar to the model chain is searched in the scene, based on a local comparison.
A Dynamic Programming-based search algorithm reduces the time complexity for the search of the model chain in the scene. Our system can find a reasonable matching, although there exist severly distorted objects in the scene. We demonstrate the feasibility of the DP-based matching method using both synthetic and real images.