| 摘要 |
<P>PROBLEM TO BE SOLVED: To perform display fit to an actual three-dimensional model or a recognition object, when performing stereoscopic vision display of the three-dimensional model correlatively with an image used for three-dimensional recognition processing. <P>SOLUTION: After recognizing a position and a rotation angle of a workpiece by recognition processing by a three-dimensional model, the three-dimensional model is subjected to coordinate transformation based on a recognition result, and a Z-coordinate after coordinate transformation is corrected corresponding to an angle (elevation angle ϕ) formed by a line-of-sight direction and an imaging surface. Then, the corrected three-dimensional model is perspectively transformed onto a coordinate system of a camera which is a processing object, and each height corresponding to a Z-coordinate before correction of each corresponding point of the three-dimensional model before coordinate transformation is set on each point of a generated projection image. A stereoscopic vision image of the three-dimensional model is generated by performing projection processing from a designated line-of-sight direction aiming at a point group distributed three-dimensionally by this processing. <P>COPYRIGHT: (C)2010,JPO&INPIT |
