Font Size: a A A

Modulation transfer function and three-dimensional shape measurement using pseudorandom structured light

Posted on:2001-11-28Degree:Ph.DType:Dissertation
University:The University of Texas at DallasCandidate:Teipen, Brian ThomasFull Text:PDF
GTID:1468390014452503Subject:Engineering
Abstract/Summary:
Over the years, several optical and opto-electronic systems have been invented in order to reproduce or improve an ability to image a physical scene. Optical imaging (vision) delivers the capability to collect an enormous amount of information about the surrounding environment. The quality of any imaging system can be conveniently defined by the amount of information that is ultimately gathered. Information transfer is the fundamental process of imaging systems and often can be described in terms of linear systems. Conveniently, transfer functions common to linear systems theory (communications theory) can be adopted to accurately describe optical information transfer in imaging systems. In this dissertation, we report on the advancements of two areas of optical imaging. Firstly, we present a new measurement process involving pseudo-random spatial patterns generated with a liquid crystal display (LCD) that allows convenient determination of the spatial-frequency transfer function commonly known as the modulation transfer function (MTF) of digital imaging systems. We present horizontal and vertical MTF measurements for color and monochrome charge-coupled device (CCD) video systems, common to machine vision and robotic vision applications. Results are found to agree with previously applied methods of measuring imaging resolution, while increasing the speed, adaptability, and convenience of the measurement. Secondly, we present three-dimensional (3-D) shape measurements from a new experimental method using polarization-encoded spatial patterns. The measurement technique is an active method of vision using projected patterns from a LCD projector. Part of the polarization-encoded pattern, a sub-pattern that is a pseudo-random code, is dedicated to solve the difficult problem of measuring the absolute height of an object with edge discontinuities. This technique conveniently allows lost fringe-order information of the classic synthetic-fringe pattern to be regained.
Keywords/Search Tags:Transfer function, Systems, Measurement, Information, Using, Optical
Related items