godawski@usceast.UUCP (04/01/87)
Hello: For quite some time, I have been working on a problem that is to locate damage caused by glaucoma. My approach is to isolate the optic disk (which can clearly be seen on the image), then the edges of the "living" optic nerve and compare these results to a newer image. I have tried many different edge detectors and many edge detection schemes, and none have worked. Thus, I am here requesting a little help. Anybody know of a good edge detector? All I need to do is to trace the outline of a "sort-of" clearly defined object, but nothing seems to want to work. Sources would be nice. Pascal sources would be particularly nice. I attempted to understand the Zero Crossing algorithm outlined in Marr's "Vision." This was a complete failure--I don't have that much of an ip background and the person I'm doing this for is breathing down my back. Any help would be greatly appreciated Thank you. Mike Godawski -- <--> Mike Godawski <--> USMail : 1907A Wheat Street; Columbia SC 29205 <--> BITNET : T170139 @ UNIVSCVM <--> ncr-sd!ncrcae-\ <--> E-mail : akgua-->!usceast!godawski <--> ihnp4-/