论文部分内容阅读
未知三维物体模型自动重建是机器视觉领域中经典研究课题之一,目前被广泛应用于工业检测、机器人定位与导航、逆向工程和文物修复等方面。首先确定出单目线激光视觉测量系统的可视区域,由此获得未知模型极限可视表面。通过将两者相结合,提出面向未知三维模型自动重建的新方法。依据初始视点下所获模型表面信息,构造未知空间的极限面模型。然后以极限面来预测未知对象的最大表面延拓信息,结合系统可视区域确定下一视点的可视性判据,给出预测曲面的可视旋转和平移区间,并将可视区间内能获取最大曲面面积的位置定为下一个最优视点的位置。通过对实体模型的自动重构,验证了方法的可行性及有效性。
The automatic reconstruction of unknown three-dimensional object model is one of the classic research topics in the field of machine vision. It is widely used in industrial inspection, robot positioning and navigation, reverse engineering and relic restoration. First, the visual area of the monocular laser vision measurement system is determined, and the unknown model limit visual surface is obtained. By combining the two, a new method for automatic reconstruction of unknown three-dimensional model is proposed. According to the model surface information obtained from the initial viewpoint, the limit surface model of unknown space is constructed. Then, the maximum surface extension information of unknown objects is predicted by the limit surface. The visibility criterion of the next viewpoint is determined according to the visible area of the system, and the visual rotation and translation intervals of the prediction surface are given. The position where the maximum surface area is obtained is set as the position of the next optimum viewpoint. Through the automatic reconstruction of the solid model, the feasibility and effectiveness of the method are verified.