Adaptive compressed photon counting 3D imaging based on wavelet trees and depth map sparse representation.

Area of Science:

• Photon counting imaging
• 3D imaging systems
• Computational imaging

Background:

• Traditional 3D imaging methods can be time-consuming and computationally intensive.
• Acquiring high-resolution 3D data often requires complex setups and significant data acquisition.
• Limitations exist in imaging through partially-transmissive occluders.

Purpose of the Study:

• To demonstrate a novel photon counting 3D imaging system.
• To develop a multiresolution technique for efficient 3D image and depth map acquisition.
• To enable 3D imaging in challenging conditions, including partial occlusions.

Main Methods:

• Utilizing short-pulsed structured illumination.
• Employing a single-pixel photon counting detector.
• Implementing a multiresolution reconstruction approach based on wavelet trees and sparse representations for depth maps.

Main Results:

• Acquisition of high-resolution 3D images and depth maps (up to 512x512 pixels).
• Significant reduction in measurement requirements and reconstruction time, with practical acquisition times as low as 17.5 seconds.
• Successful imaging through partially-transmissive occluders and direct acquisition of novelty images for change detection.

Conclusions:

• The proposed photon counting 3D imaging technique offers an efficient solution for high-resolution 3D scene reconstruction.
• The method's speed and ability to handle occlusions make it suitable for dynamic scenes and challenging imaging environments.
• This approach advances computational imaging by enabling rapid, detailed 3D scene analysis.