Calibration method of spatial transformations between the non-orthogonal two-axis turntable and its mounted camera

Area of Science:

• Robotics and Automation
• Computer Vision and Image Processing
• Mechanical Engineering

Background:

• Cameras mounted on two-axis turntables are crucial for expanding field of view and measurement range in visual tasks.
• Accurate calibration of camera position and attitude relative to the turntable is essential for reliable visual measurements.
• Conventional methods assume ideal orthogonal turntables, failing to account for real-world deviations like non-intersecting axes or off-center camera mounting.

Purpose of the Study:

• To propose a novel position and attitude calibration method for cameras mounted on non-orthogonal two-axis turntables.
• To address the significant errors caused by discrepancies between actual physical turntables and idealized models.
• To establish a robust calibration technique that accounts for non-ideal turntable geometries.

Main Methods:

• The method models the spatial relationship between the azimuth and pitch axes of the turntable, even when they are not orthogonal or intersecting.
• It leverages the geometric invariant characteristics of the camera during motion.
• Turntable axes are recovered, a base coordinate system is established, and subsequently, the camera's position and attitude are calibrated.

Main Results:

• The proposed method accurately describes the spatial relationship between non-orthogonal turntable axes.
• It successfully recovers turntable axes and establishes a base coordinate system.
• Calibration of the camera's position and attitude is achieved, validated by simulations and experiments.

Conclusions:

• The developed calibration method is effective for non-orthogonal two-axis turntables, a common scenario in real-world applications.
• It significantly improves the accuracy of visual measurements by accounting for real-world turntable imperfections.
• The method provides a reliable solution for camera-turntable calibration, enhancing the performance of various visual tasks.