A robust system calibration method for the fringe projection profilometry

Konferenz: MEMAT 2022 - 2nd International Conference on Mechanical Engineering, Intelligent Manufacturing and Automation Technology
07.01.2022 - 09.01.2022 in Guilin, China

Tagungsband: MEMAT 2022

Seiten: 4Sprache: EnglischTyp: PDF

Autoren:
Li, Wenjie; Zhang, Zonghui; Li, Tuanxing; Huang, Yang; Jiang, Zhansi; Gao, Xingyu (School of Mechanical and Electrical Engineering, Guilin University of Electronic Technology, Guilin, China)
Tan, Zhengdong (Shenzhen Anewbest Electronic Technology Limited Company, Shenzhen, China)
Li, Tengfei; Huang, Meikuan (Zhuhai Huaya Machinery Technology Limited Company, Zhuhai, China)

Inhalt:
With the advantages of non-contact, simple structure, high precision and dense data, the fringe projection profilometry (FPP) is widely used in the product inspection and measurement. The system calibration is the key step for the FPP, and the eight parameters calibration method is popular to the single camera-projector system. The information of feature points, including the pixel coordinates, unwrapping phases and camera coordinates, is used to get the calibration parameters. However, the information is suffered from the noise easily. To enhance the robustness of system calibration, a noise suppression method based on the random sample consensus (RANSAC) plane fitting is proposed. Compared with the least square based plane fitting, the noise points are not used to fit the plane in the RANSAC based plane fitting, which can get a higher precision. The efficiency and feasibility are both verified in theory and experiment analysis.