Type of Document	Master's Thesis
Author	Tatem, James E
URN	etd-05092009-040743
Title	Automatic image analysis methods for use with local operators
Degree	Master of Science
Department	Electrical Engineering
Advisory Committee	Advisor Name Title Nadler, Morton Committee Chair Conners, Richard W. Committee Member Ehrich, Roger W. Committee Member
Keywords	Computer graphics.
Date of Defense	1990-05-01
Availability	unrestricted
Abstract Just as image processing and image data bases have moved out of the lab and into the office environment, so has the need for image enhancement. Image scanners must to be able to capture and store a wide variety of information including faded documents, carbon copies, signatures, postmarks, etc. OCR systems put further demands on scanned image quality in terms of low noise, and unbroken disconnected characters. Straight thresholding techniques do not always meet the performance requirements, but by applying simple image processing techniques some of these problems can be solved. However, more burden is placed on the users to control the image enhancement techniques. The users, most of whom have little technical background, want no part in adjusting parameters. This paper proposes a method of examining small windows of the image to derive parameter settings autonomously. Histograms allow rudimentary measures to be used in setting parameters for edge detection, non-linear filters, and point operators such as non-linear gray scale mapping. Some examples of automatic parameter setting are given in chapter three.
Files	Filename       Size       Approximate Download Time (Hours:Minutes:Seconds)     28.8 Modem   56K Modem   ISDN (64 Kb)   ISDN (128 Kb)   Higher-speed Access    LD5655.V855_1990.T384.pdf 3.22 Mb 00:14:53 00:07:39 00:06:41 00:03:20 00:00:17 next to an author's name indicates that all files or directories associated with their ETD are accessible from the Virginia Tech campus network only.

If you have questions or technical problems, please Contact DLA.