Modeling and calibration of a precise optical positioning system based on four linear cameras

Area of Science:

• Optical Engineering
• Metrology
• Computer Vision

Background:

• Precise 3D positioning systems are crucial for applications in large spaces.
• Existing systems using linear cameras face challenges with cylindrical lens distortion.
• Calibration of linear cameras with cylindrical lenses is a long-standing problem impacting accuracy.

Purpose of the Study:

• To develop and calibrate a precise optical positioning system for 3D remote target tracking.
• To address and overcome the distortion issues inherent in cylindrical lenses used in linear cameras.
• To improve the accuracy and efficiency of 3D position tracking in large-scale environments.

Main Methods:

• A system comprising four linear cameras with cylindrical lenses, arranged in two paired groups.
• Each camera group functions as an integrated 2D image sensor with orthogonal imaging orientations.
• Modeling and calibration using object-space error to determine optimal camera parameters for each group.

Main Results:

• The proposed system effectively eliminates distortion from cylindrical lenses.
• The calibration approach successfully fits the linear camera model and corrects lens distortion.
• Simulative and real experiments demonstrate significant improvements in positioning accuracy.

Conclusions:

• The novel paired linear camera system offers superior performance for 3D positioning.
• Effective calibration and distortion correction lead to enhanced tracking accuracy.
• This method provides a robust solution for precise remote target tracking in large spaces.