Computational optical distortion correction using a radial basis function-based mapping method.

Area of Science:

• Optics and Photonics
• Computer Vision and Image Processing
• Human-Computer Interaction

Background:

• Off-axis head-worn displays (HWDs) often exhibit significant asymmetric optical distortion.
• Accurate image rendering is crucial for effective visual immersion and user experience in HWDs.
• Existing distortion correction methods may lack the precision required for high-fidelity HWD applications.

Purpose of the Study:

• To present a distortion mapping and computational image unwarping method for off-axis HWDs.
• To achieve high accuracy in correcting asymmetric distortion across the display's field of view.
• To validate the efficacy of the proposed unwarping method through experimental assessment.

Main Methods:

• Utilizing a network interpolation approach based on radial basis functions (RBFs) for distortion mapping.
• Applying the method to correct up to 23% asymmetric distortion in an off-axis HWD with a 27°x21° field of view.
• Assessing unwarping efficacy using optical design software, correlation coefficients, and edge superimposition.

Main Results:

• Achieved a mapping function with a 10⁻⁵ mm absolute error across the field of view.
• Demonstrated successful prewarping of images using RBFs for a specific off-axis HWD.
• Generated real-time video with minimal latency (2 ms) using a custom application.

Conclusions:

• The presented RBF-based distortion mapping and unwarping method effectively corrects significant asymmetric distortion in off-axis HWDs.
• The method achieves high accuracy, enhancing the visual quality of images displayed on HWDs.
• The successful real-time implementation demonstrates the practical applicability of this technique for immersive HWD systems.