Improving the noise immunity of 3D computational ghost imaging.

Area of Science:

• Optics and Photonics
• Computational Imaging
• 3D Reconstruction Technologies

Background:

• Computational ghost imaging (CGI) reconstructs 3D images from shading information.
• 3D image reconstruction quality in CGI is susceptible to noise and initial position selection.
• Existing methods struggle with noise and optimal starting point determination.

Purpose of the Study:

• To enhance the accuracy and quality of 3D images reconstructed using CGI in noisy environments.
• To introduce a novel method for selecting the optimal initial growing position in CGI.
• To improve the robustness of CGI against noise during 3D reconstruction.

Main Methods:

• Application of sub-pixel displacement technique for smoothing shading images.
• Development of an optimal initial growing position selection strategy.
• Comparative analysis of the proposed method against existing CGI techniques.

Main Results:

• Achieved smoother shading images in noisy environments.
• Successfully preserved stereo features of the object through optimal initial position selection.
• Demonstrated more accurate 3D image surface reconstruction compared to previous methods.

Conclusions:

• The proposed method significantly improves 3D image accuracy in CGI, especially under noisy conditions.
• Sub-pixel displacement and optimal initial position selection are crucial for robust CGI.
• This research advances the application of CGI for 3D imaging in challenging environments.