FAQ

For Berkeley Image Seg updated on February 13, 2012.

Contact support@BerkEnviro at any time to get in touch with James.

Q: How are the fields in the stats CSV file defined?

Please see this Google Doc spreadsheet. Some formulas are reimplemented from FRAGSTATS.

Thanks to Matt Stevenson at CORE GIS for kicking this off.

Q: I just downloaded and installed the trial of Berkeley Image Seg, why is it already complaining about "30 days trial has expired / cannot locate license file"?

Sorry, this is a shortcoming of our installer. If BIS installed itself to a directory other than exactly "C:\Program Files\BerkeleyImageSeg\", please make a directory called "C:\Program Files" and copy the BIS installation directory into it.

Do not move or delete the original install, those executables and shortcuts should now work as expected. Copying the files is a workaround that should simply let the license module work.

This is an issue on non-English systems using a different name for "Program Files". Also on a 64-bit OS BIS will install to "C:\Program Files (x86)", which won't fly.

Thanks to Paulo from Italy and 64-bit Mike for testing this.

Q: What is the difference between the GUI Wizard and command line versions?

The Wizard is a "demonstrator GUI" that guides a new user through the various settings and output options. The command line interface is what you want to settle into for efficient expert use and automation. Type “BerkeleyImageSeg.exe ?” for command line options.

Functionally, the command line does not support classification but does support parallel processing and will manage a large number of batch runs. The GUI only supports a limited amount of batch runs. In summary, it is the command line version you want to run for segmentation batching albeit without classification.

Q: I got the error message that the image is too big for further processing. How do I process an image this size?

The command line exe has an +x option that will tile the processing. With the +p option it will also dispatch those tiles to multiple cores/processors. The Wizard cannot tile an image, rather it is good for testing smaller clips. Type “BerkeleyImageSeg.exe ?” for command line options.

We highly recommend finding the best segmentation parameters on a small clip of your image, say 500x500 or 1000x1000 pixels. That way it is feasible to run dozens or even thousands of parameter combinations. Choose a representative sub-region of your image and run the iterations as described below. Then on the handful of most suitable parameters, run those on the full image.

Q: What do the parameters threshold, shaperate, and compactness mean and do?

In essence, threshold determines how large the segments get. Shaperate and compactness are weights that determine how the shapes look. Shaperate weighs the shape attributes versus the color attribute. Then within the shape calculation, the compactness rate weighs the compactness calculation over the smoothness calculation.

The main thing to control: A lower shaperate will let the segments go out further to follow similar colors. A higher shaperate will generally keep segments closer in, albeit less spectrally homogenious.

For instance a threshold of 50, shaperate of 0.7, and compactness rate of 0.5 says "Give the regions 50 growth cycles while weighing the shape calculations over color homogeneity by 70/30 and (within the 70 percent bias) give equal weight to the compactness and smoothness calculations."

Q: How do I choose the parameters? Better yet, how do I just segment an image and get a result that I like?

The short answer: Run all possible segmentation combinations and choose the one that looks the best! That is how Berkeley Image Seg is designed: Let the computer do the number crunching and you choose what output looks best. The training and ranking functions can assist.

It is best to do this computationally intensive step on a 1000x1000 or smaller representative clip of your image.

Q: I need more powerful classification tools. What do you recommend?

We recommend putting together a workflow of Berkeley Image Seg plus the open source Weka project (GPL).

Q: Will BIS run on Unix/Linux, can it plug into ArcGIS, can I use Python directly?

Yes. The underlying source code is in Python/Pyrex. The Python API can be exposed or compiled to another platform by request. Please get in touch with James to voice your interest in a particular platform.

Q: Can I incorporate Berkeley Image Seg into a larger workflow?

Yes. As a stand alone windows program, the command line interface is batchable. BIS works well as a "segmentation kernel" between third party image pre-processing, spatial statistics, and object classification software.

Q: Can I embed the technology behind Berkeley Image Seg in my own application?

Sure. Get in touch with James with your idea.

Q: Where does the Berkeley in Berkeley Image Seg come from?

The scientists who developed the software have all worked at and/or earned a degree at UC Berkeley. It's how we know each other. The company or product does not have any formal tie to the university.