Refraction correction for deep-water three-dimensional visual measurement based on multi-objective optimization.

Area of Science:

• Optical Engineering
• Robotics
• Marine Technology

Background:

• Refraction significantly impacts the accuracy of 3D visual measurements in deep-water settings.
• Existing methods struggle to precisely calibrate for refractive distortions in underwater environments.

Purpose of the Study:

• To develop a high-accuracy binocular camera refractive imaging model for deep-water 3D measurements.
• To propose a calibration method for refraction parameters to enhance shape and deformation analysis.

Main Methods:

• Established a binocular camera refractive imaging model.
• Utilized a 3D calibration target for initial refractive axis estimation.
• Employed the Non-dominated Sorting Genetic Algorithm II (NSGA-II) for optimizing refraction parameters using dual objectives (spatial point pair distance and reprojection errors).
• Developed an improved numerical method to accelerate analytical forward projection for efficient reprojection error calculation.

Main Results:

• Achieved an average absolute position error of less than 1.1 mm for reconstructed points.
• Attained an average displacement error of less than 0.04 mm.
• Demonstrated the method's effectiveness through underwater experiments.

Conclusions:

• The proposed method provides a robust solution for accurate 3D visual measurements in deep-water environments.
• The calibration technique effectively mitigates refraction-induced errors, enabling precise shape and deformation analysis.