Subjective speckle suppression for 3D measurement using one-dimensional numerical filtering.

Area of Science:

• Optics and Photonics
• Computer Vision and Image Processing
• Metrology

Background:

• Laser speckle projection is vital for 3D reconstruction in stereo photogrammetry.
• Coherent noise, or subjective speckles, degrades image quality during surface imaging.
• Existing noise reduction techniques are often time-consuming or reduce light intensity and depth of field.

Purpose of the Study:

• To introduce numerical 1D filtering techniques for reducing subjective speckles in laser speckle projection.
• To enhance the performance of 3D reconstruction without altering experimental setups.
• To investigate the impact of filter orientation on noise reduction and reconstruction accuracy.

Main Methods:

• Development of a model for contrast reduction analysis.
• Investigation of the relationship between filter orientation and setup geometry.
• Implementation and evaluation of 1D filtering through simulations and real-world experiments.

Main Results:

• Numerical 1D filtering significantly reduces coherent noise in laser speckle patterns.
• Performance improvements in 3D reconstruction were observed, particularly with optimized filter parameters.
• A vertical filter orientation relative to the setup geometry was found to be generally most effective.

Conclusions:

• Numerical 1D filtering offers a practical solution to subjective speckles in laser speckle projection.
• This approach enhances 3D reconstruction accuracy and efficiency without requiring experimental modifications.
• Filter orientation is a critical parameter for maximizing noise reduction and reconstruction quality.