Towards Spatially-Selective Lensing for Imaging and Vision

Abstract

A core component of any imaging system is the lens, which gathers bundles of light to form images. These images appear sharp when objects are within the lens depth of field, whose range is inversely related to the lens light gathering ability. To form sharp images of scenes spanning a large range of depths, one must typically extend the depth of field by shrinking the lens aperture at the cost of reducing light levels. This project investigates a novel approach to all-in-focus imaging that circumvents this fundamental trade-off, through the use of a spatially-selective computational lens. This lens uses a combination of a Lohmann lens and phase spatial light modulator (SLM) to provide perpixel programmable control of focal length. By estimating the depth at every pixel, the lens can be programmed to bring a dense set of focal planes into focus all at the same time. The objective of this project is to advance the theory, design, and applications of a spatially-selective lens to imaging systems. This includes providing an analysis of the imaging limits of spatiallyselective lensing, investigating spatially-adaptive autofocus algorithms to program the lens, and designing phase masks to enable other operations that go beyond focusing. The proposal targets applications that span a gamut of scene ranges and sizes, including imaging remote targets with a telephoto lens or microscopic scenes.

Document Details

Document Type
DoD Grant Award
Publication Date
Feb 06, 2025
Source ID
FA95502410244

Entities

People

  • Matthew O Toole

Organizations

  • Air Force Office of Scientific Research
  • Carnegie Mellon University
  • United States Air Force

Tags

Fields of Study

  • Physics

Readers

  • Computer Vision.
  • Image Processing and Computer Vision.