How to Go From Raster to Vector After Edge Detection

Share icon

Thu, 06/27/2013 - 11:31 am

I'm new to Igor, so I apologize in advance if I ask any redundant questions or use improper terminology. I'm currently working on a project that requires we detect the edges of some structures we are imaging. Once the edges are detected, we need to export that file as an .iges to perform further analysis using a different program.

I haven't discovered any option in Igor to export to .iges. Instead, I've resorted to exporting the image from Igor as a .jpg, importing it into Illustrator to vectorize the image after performing an image trace, then importing it into Autocad in order to export it as an .iges. As is obvious, the number of programs this involves results in a laborious and time-consuming process.

My first question, then, is whether or not Igor can export directly to .iges. If not, can Igor vectorize a bitmapped image?

If the answer to both these questions is "No", I would greatly appreciate any advice concerning how to go from an .ibw to an .iges.

Many thanks.
I'm not aware of any support for iges format in IGOR.

From casual reading of the format specs I get the impression of a very simple output format that you could probably write (for simple shapes) using the Open, fprintf and Close commands.

The main difficulty is going to be in converting your "edges" into vectors. In general, edge detection algorithms do not produce simple edges so unless you have a very specific and simple structure it is not an easy task. One trick that may be helpful here is to take the output of the edge detection and pass it through ImageAnalyzeParticles in order to obtain the boundary waves which are a vector form representation of the edges.

A.G.
WaveMetrics, Inc.

June 27, 2013 at 12:10 pm - Permalink

Depending on what you mean by "vectorize a bitmapped image", you may want to take a look at the ImageSnake operation. There is an experiment called snakeDemo among the example experiments that illustrates what the ImageSnake operation does.

June 30, 2013 at 04:27 am - Permalink

Thank you both very much. I have explored both the ImageAnalyzeParticles function and the snakeDemo example experiment over the past few days and have come to a much greater understanding of the subject as a result. While I still have some questions on the topic, these replies have helped point me in the right direction.

July 3, 2013 at 07:46 am - Permalink