C3 Lensless: Computation and Communication enabled on-Chip Lenseless Imager

Abstract

Cameras are becoming ubiquitous and have found innumerable DOD-relevant applications. While traditional lens-based imaging devices have served many of these applications faithfully, several emerging DOD applications such as surveillance using micro-drones and disaster recovery require image sensors to be compact, light-weight, power-efficient, and inexpensive. Existing lens-basedcameras remain constrained in terms of their size, weight, and cost. The primary reason for these constraints is the lens -- which typically accounts for more than 90% of the size, weight and cost of these cameras. Our group at Rice University has pioneered the design and development of FlatCam -- a lensless imaging system that makes order of magnitude reductions to the size, weight and cost of imaging by completely eliminating the need for a lens. In addition, many of theseemerging applications also require on-chip intelligence and communication capabilities -- so that they are able to detect and identify targets and make on-field decisions.The power requirements of state-of-art image sensing and associated machine learning algorithms make these impractical on battery-operated devices. FlatCam by combining optical computing (which is extremely power efficient), with heavily optimized on-chip machine learning algorithms, can reduce the power requirements by two orders of magnitude enabling intelligent, light-weight,compact and inexpensive imaging. With these goals in mind, we will use this DURIP award to fabricate, test and characterize C3 (Computation, and Communication enabled on-Chip) Lensless imaging devices. If realized, these advances would revolutionize imaging and situational awareness in the battlefield, providing a significant strategic benefit to our military. The equipment provided from this DURIP will be integrated with current research infrastructureto create a streamlined system to develop and characterize miniature C3 Lensless imaging devices. Combining these on-chip imaging devices with the rich mathematical tools of machine learning and optimization will fuel a new research field based on co-designing hardware and algorithms to produce novel imaging capabilities at reduced SWAPC constraints. In addition, the proposedinstruments will greatly enhance both undergraduate and graduate education/research at Rice University by exposing a new generation of students to this highly interdisciplinary research.Furthermore these chip-scale devices could have important economic impact by penetrating thecommercial markets for imaging devices in consumer imaging, medical imaging, bio-imaging, security, astronomy and other applications.

Document Details

Document Type
DoD Grant Award
Publication Date
Jun 13, 2019
Source ID
N000141912440

Entities

People

  • Ashok Veeraraghavan

Organizations

  • Office of Naval Research
  • Rice University
  • United States Navy

Tags

Readers

  • Distributed Systems and Data Platform Development
  • Image Processing and Computer Vision.
  • Integrated Circuit Design and Technology.

Technology Areas

  • AI & ML
  • AI & ML - DoD AI Strategy
  • Autonomy
  • Fully Networked C3
  • Fully Networked C3 - Command and Control