A High-Resolution 3-D Shape Measurement System
In this paper, we propose a two-dimensional sub-pixel interpolation method in the process of stereo matching to enhance the accuracy of the combined stereovision and phase shifting method for 3-D shape measurement. In the original combined stereovision and phase shifting method, the left and right images were matched at the pixel level. Since it is impossible to find two pixels with exactly the same horizontal and vertical phase values, significant digitization errors were present. In this paper, we describe a two-dimensional sub-pixel interpolation method used to match the left and right images at the sub-pixel level, which greatly reduces the system error. To reduce the random noise, we also collect multiple sets of images and calculate their average. Experimental results showed a RMS error of 11 fan in a field-of-view of approximately 100.0 X 66.7 mm after the proposed methods were implemented. The effectiveness of our proposed methods has been experimentally confirmed. .
sub-pixel interpolation stereo matching 3-Dshape measurement stereovision phase shifting random noise
Kairen Deng Peisen Huang
State Key Laboratory of Mechanical Systems and Vibration UM-SJTU Joint Institute Shanghai Jiao Tong University Shanghai, China
国际会议
重庆
英文
108-112
2011-01-21（万方平台首次上网日期，不代表论文的发表时间）