DoF-Dependent and Equal-Partition Based Lens Distortion Modeling and Calibration Method for Close-Range...

Area of Science:

• Computer Vision
• Optical Metrology
• Photogrammetry

Background:

• Lens distortion significantly impacts close-range vision measurement accuracy, particularly concerning the depth of field (DoF).
• Existing methods often struggle with DoF-dependent distortions, especially in short-distance, small-focal-length photography.
• Accurate calibration of these distortions is crucial for reliable 3D reconstruction and measurement.

Purpose of the Study:

• To propose a novel lens distortion modeling and calibration method that accounts for depth of field (DoF) dependency.
• To address the limitations of current models in short-distance and small-focal-length photography.
• To enhance the accuracy of close-range vision measurements by precisely characterizing DoF-dependent lens distortions.

Main Methods:

• A DoF-dependent distortion model was developed, independent of manual focus adjustments, simplifying the calibration process.
• A 2D-to-3D equal-increment partitioning method was introduced to improve distortion representation accuracy perpendicular to the optical axis.
• A calibration control field was designed, enabling de-coupled calibration of distortion and other camera parameters by analyzing extracted line segments.

Main Results:

• The proposed equal-increment partition-based DoF distortion model achieved maximum projection errors of 0.11 pixels and average errors of 0.05 pixels.
• Angular reconstruction errors were found to be as low as 0.013° (maximum) and 0.011° (average).
• Experimental results validated the effectiveness and accuracy of the developed lens distortion model and calibration technique.

Conclusions:

• The proposed DoF-dependent lens distortion model and calibration method are effective for close-range photography.
• The method significantly improves the accuracy of vision measurements by accurately characterizing depth of field-dependent distortions.
• This work provides a robust solution for precise camera calibration in applications requiring high accuracy at short distances.