Computer-generated holography encoding via dynamic region segmentation and parallel phase difference inference.

Area of Science:

• Optical Engineering
• Metrology
• Computational Optics

Background:

• Freeform optical elements are crucial for high-performance imaging systems.
• Computer-generated holograms (CGHs) for null testing face challenges in encoding accuracy, computation, and data size due to complex fringe patterns.
• Efficient CGH encoding is vital for the widespread adoption of freeform optics.

Purpose of the Study:

• To develop a novel CGH encoding method for freeform optics null testing.
• To address the trade-offs between encoding accuracy, computational workload, and data size.
• To improve the efficiency and practicality of CGH encoding for complex optical elements.

Main Methods:

• Leveraging fringe pattern structural characteristics and spatial phase continuity to infer phase values.
• Theoretical analysis comparing quadratic growth of conventional methods to linear growth of the proposed method.
• Application and verification of the encoded CGH in interferometric testing with error modeling.

Main Results:

• The proposed method achieves linear growth in explicit points, unlike the quadratic growth of conventional methods.
• Demonstrated significant reductions in runtime (approx. 107.92x) and data size (approx. 99x) for a large-diameter CGH.
• Maintained encoding error below λ/200, validating high accuracy.
• Successful application in interferometric testing confirmed accuracy and effectiveness.

Conclusions:

• The developed CGH encoding method offers a favorable balance between encoding accuracy, computational efficiency, and storage demand.
• This approach provides a practical solution for high-precision CGH encoding in freeform optics null testing.
• The method facilitates the efficient and widespread use of freeform optics in advanced imaging applications.