摘要
分析基于Marr算子的标定算法存在易受环境光及噪声影响、精度不够高的不足 ,根据计算机视觉标定的目标要求 ,通过动态阈值分割、边缘提取、轮廓跟踪和两步法标定算法等处理策略 ,克服了传统标定算法的缺点 ,并实现了计算机视觉摄像机标定的整个过程自动化 .实验结果证明 ,该标定算法鲁棒性强 ,操作简便 ,抗噪声、抗环境光性能好 。
According to the requirements of the calibration of computer vision, a series of effective steps including dynamic threshold dividing, contour picking, contour tracing, and two stage calibration algorithm; and the whole process automation of the calibration were achieved. The result showed that this calibration algorithm was characterized by high robustness, easy operation, non noisesensitivity, and high-accuracy.
出处
《华中科技大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2004年第8期13-15,共3页
Journal of Huazhong University of Science and Technology(Natural Science Edition)
关键词
计算机视觉
标定
轮廓跟踪
自动化
computer vision
calibration
contour tracing
automation
