Study for the introduction of 3D compressed sensing mechanism in a classical LIDAR hardware

PDF

Authors
  1. Tremblay, G.
Corporate Authors
Defence Research and Development Canada, Valcartier Research Centre, Quebec QC (CAN);AEREX Avionics Inc, Breakeyville QUE (CAN)
Abstract
The development of the micro mirrors device (DMD) technologies paved the way to numerous new applications. One of the most unexpected is compressed sensing imaging using single detector. To produce a 2D or 3D image using compressed sensing (CS), one needs to focus the image of a scene on a modulator (in our case, a DMD). By changing the modulation of the DMD, we create a system of n equations with n unknowns. This kind of system is usually difficult to solve unless the scene can be easily compressed. We made tests using the GPSR 6.0 algorithm to reconstruct simulated 3D scene. These tests showed that high resolution image reconstruction requires multiple acquisitions and processing iterations. In our tests, the whole process could take up to 1 s for a single low resolution frame. This tends to demonstrate that real time high resolution 3D imaging is not possible with the current state of the technology and single-pixel systems. However, the technology is less expensive than the alternative ICCD cameras of Geiger-mode APD matrix. Considering the cost advantage, application such as monitoring and surveillance could benefit from this technology.

Il y a un résumé en français ici.

Keywords
Lidar;3D compressed sensing
Report Number
DRDC-RDDC-2017-C069 — Contract Report
Date of publication
01 Nov 2016
Number of Pages
35
DSTKIM No
CA044877
CANDIS No
805252
Format(s):
Electronic Document(PDF)

Permanent link

Document 1 of 1

Date modified: