Repository logo
Log In(current)
  1. Home
  2. Colleges & Schools
  3. Graduate School
  4. Masters Theses
  5. A microcomputer-based electronic radiographic inspection system using digital image processing techniques
Details

A microcomputer-based electronic radiographic inspection system using digital image processing techniques

Date Issued
June 1, 1986
Author(s)
Younkin, James Russell
Advisor(s)
Donald W. Bouldin
Additional Advisor(s)
Dragana Brzakovich
Robert Rochelle
Permanent URI
https://trace.tennessee.edu/handle/20.500.14382/35416
Abstract

The purpose of this task was to provide the hardware and software requisite to implement an electronic radiographic workstation that would allow the suitability of both image processing techniques and imaging components to be evaluated as candidate solutions for Non-destructive evaluation requirements in an industrial environment. The workstation is based around the LSI-11/73 microcomputer and its Q-bus architecture and the single-user RT-11 operating system. An image processor and array processor have been integrated into the host computer system. Specific software was developed to provide a reasonable performance, low cost, workstation. Workstation features include expandable mass storage, hard copy image output, M Mbyte of accessible array storage on the host computer, direct frame buffer access to data by the array processor, continuous histogram display, and frame buffer arithmetic operations. As a culmination to this study, the workstation and image processing techniques are used as an automated solution to the problem of detecting and measuring foreign inclusions within a material before it is mechanically processed. The steps involved toward automating this inspection include the acquisition of the image, preprocessing the image, thresholding it with a global threshold, and locating and measuring the impurities. The preprocessing step is composed of a non-linear filtering algorithm that reduces the noise in the image and preserves the defect's edge information and a polynomial-fit-subtraction technique to eliminate a slowly varying background gradient present in the acquired images. These two techniques put the image data into a more suitable form to permit further automated processing. The effects of the processing step allows the area of 0.020 inch diameter test spheres to be measured within 2.5% by computer measurement.

Degree
Master of Science
Major
Electrical Engineering
File(s)
Thumbnail Image
Name

Thesis86Y685.pdf

Size

7.43 MB

Format

Unknown

Checksum (MD5)

24bdf76a4efb4332d03d38d5f9f53d0a


University Libraries

1015 Volunteer Boulevard
Knoxville, TN 37996
865-974-4351

Map & Directions
Donate to the Libraries
  • About
  • John C. Hodges Society
  • Speaking Volumes magazine
  • Outreach
  • Directory
  • Employment
  • Policies
  • Library Intranet
University of Tennessee power T logo

The University of Tennessee, Knoxville
Knoxville, Tennessee 37996
865-974-1000

Events
A-Z
Apply
Privacy
Map
Directory
Give to UT
Accessibility

Built with DSpace-CRIS software - Extension maintained and optimized by 4Science