论文部分内容阅读
为了实现气浮实验台位姿的实时确定,提出了一种基于单目视觉的位姿测量方法。首先,设计了一种具有旋转、平移、缩放不变性且易于探测的圆形合作靶标;其次,结合靶标尺寸、形状以及安装位置,基于Blob分析快速识别合作靶标,保证了定位点提取的准确性;然后,提出一种在待识别区域内进行“行、列”扫描统计的靶标圆心测量算法,实现靶标圆心的快速、鲁棒提取;最后,结合计算机坐标系下的圆心位置以及视觉测量系统中坐标变换关系,解算气浮实验台的实际位姿。实验结果表明:该方法测量位姿精度较高,抗噪能力强,在处理1 600 pixel×1 600 pixel图片的情况下,平均测量周期为53.086 ms(约19帧/s),能够实现对气浮实验台位姿实时、精确、鲁棒的测量。
In order to determine the position and attitude of the flotation experiment platform in real time, a pose measurement method based on monocular vision is proposed. Firstly, a circular cooperative target with rotation, translation, scaling invariance and easy detection is designed. Secondly, combining with the target size, shape and installation position, the target is quickly identified based on Blob analysis to ensure the accuracy of the location extraction Then, a new algorithm called “line and column” scanning centering around the area to be identified is proposed to realize the fast and robust extraction of the center of the target. Finally, combining the position of the center of the circle in the computer coordinate system and the visual measurement System coordinate transformation relationship, the calculation of the actual posture of the air flotation bench. The experimental results show that the proposed method has high attitude and attitude accuracy and strong anti-noise ability. The average measurement period is 53.086 ms (about 19 frames / s) in the case of 1 600 pixel × 1 600 pixel picture, Floating experimental position pose real-time, accurate and robust measurement.