基于SiftGPU特征匹配方法的实时视觉里程计系统

摘要

The visual odometry method is adopted based on Scale Invariant Feature Transform(SIFT) to get stereo corresponding points for motion estimation between every two binocular frames. Then the rotation and translation matrix is employed to invert the moving path of the robot. SiftGPU is applied to accelerate computing so that real-time visual odometry system can be achieved. RANSAC algorithm is adopted to eliminate mismatching interference. Experimental results show that SIFT algorithm with strong scale and affine transform invariance can get more accurate path inversion results; and GPU graphics acceleration will achieve real-time visual positioning.%采用尺度不变特征变换(SIFT)特征匹配方法对双目相机图像进行立体匹配,同时匹配相邻两时刻的三维点,求解运动方程进行运动估计,得到机器人 2 个时刻坐标变换的旋转和平移参数;使用每 2个时刻的旋转和平移结果进行机器人的路径反演,采用 GPU加速 SIFT特征提取与匹配,实现实时的视觉里程计系统,并采用 RANSAC 算法用于运动估计剔除误匹配点干扰.实验结果表明,具有仿射变换较强不变性的 SIFT特征匹配算法能够得到较为精确的路径反演结果,采用 GPU加速 SIFT特征提取与匹配能达到实时的视觉定位效果.