1-310 モデルベーストマッチングを用いた 3-D 位置/姿勢計測方法の提案と評価

概要

論文の詳細を見る
This paper presents a vision-related technique for 3-D position/orientation measurement. The proposed method utilizes the genetic algorithm (GA) and unprocessed gray-scale image input from stereo-vision, in order to perform recognition of a target being imaged with known target object shape. In GA process, computation of fitness function is based on the configuration of an object model designated as surface-strip model in this presentation, and the problem to recognize the target shape and simultaneous detection of the position/orientation, is converted to an optimization problem of the fitness function. To verify the effectiveness of the proposed image recognition method, we have conducted experiments to detect the target position and orientation in 3-D space while recognizing its shape using static image.
社団法人日本機械学会の論文
2002-11-14