No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

The problem of recovering a wavefront or surface deviation from a fringe pattern image is investigated. The data sets generated in sampling a fringe pattern are considered in the frequency domain and errors that arise in analyzing the pattern are given an aliasing interpretation. Uniform sampling is shown to be superior to other sampling schemes. A variety of techniques for locating fringe centers are reviewed and three synchronous convolution methods are demonstrated that do not require the location of fringe centers. For the latter, direct phase measurements are obtained from fringe pattern photographs over a uniform grid with an accuracy comparable to temporal heterodyne interferometry. Two microprocessor-based video systems are described in which the traditional and synchronous convolution methods were implemented.

An electronic display is typically the last stage of an imaging chain before the human observer. Some of its physical properties can make it a bottleneck in the chain. This dissertation is an attempt to study and cope with certain critical properties of electronic displays, specifically spatial resolution, noise and contrast. The focus of this study lies in two parts. One is the measurement and estimation of display properties, including the spatial resolution and the spatial noise. The other is on the development of display-based image processing algorithms, which can compensate for the degradations from the specific display. The basic idea is to perform operations on the digital image before it is displayed, which will counteract the degradations that come from the action of the display systems. The properties of the display will be derived mainly from the digital images taken by a high quality CCD camera on the display. Based on the display properties estimated, several image processing techniques have been developed in this project to optimize the spatial resolution, noise and contrast resolution of the specific display under consideration. We have also developed a novel error-diffusion-based algorithm for adaptive image contrast enhancement which incorporates the properties of the display as well as the properties of the image.

This dissertation describes the design and implementation of a computer vision based varve image analysis system. The primary issues covered are software engineering design, varve image calibrations, varve image enhancement, varve Dynamic Spatial Warping (DSW) profile generation, varve core image registration, varve identification, boundary identification and varve thickness measurement. A varve DSW matching algorithm is described to generate DSW profile and register two core images. Wavelet Multiple Resolution Analysis (MRA) is also used to do the core image registrations. By allowing an analyst to concentrate on other research work while the VARVES software analyzes a sample, much of the tedious varve analysis work is reduced, and potentially increasing the productivity. Additionally, by using new computer vision techniques, VARVES system is able to do some varve analysis which was impossible handled manually.

The problem approached in this thesis is how to help the novice do image processing with a rule-based system. An interactive expert system prototype, which includes the image processing functions of SADIE, has been created. A software interface between the image processing environment, SADIE, and the expert system tool, CLIPS, has been coded in standard C. Knowledge about low-level image processing (contrast enhancement and noise suppression) has been successfully implemented in the CLIPS program. The experimental results have shown that this system is efficient and friendly. Although this system was developed on a VAX under VMS, it is easily modified for most machines because it includes the portability of CLIPS. With only a minimal amount of work, we also can incorporate the system with the new SADIE C version, which has been developed in another project.