I took a photograph of a raven when we were at the Grand Canyon in April 2008 (see image below). I didn't really like the photograph too much, but it was a nice template for trying to create a vector illustration along the lines of what is shown here:
This page contains a listing of all my posts. You can filter them by broad topic using the submenus under "Creative Outbursts". Or you can filter articles with the tag cloud in the sidebar to the right. You can also sign up to my RSS feed to be notified of new articles.
A modular system of interlocking pieces for holding small gadgets in place. Printed on the Daniel Lab's uPrint 3D Printer to fit a multitude of holding tasks. Standard 1/4-20 screws can be used to fasten together individual pieces. (Fall 2009 - 2010)
Notes on the design and construction of a (relatively) cheap wind tunnel for the Daniel Lab. (Summer/Fall 2009)
Using food-grade silicone, I created a few 3D molds of moth pupae, which I then used to make tasty chocolate treats for Halloween. The photos in this post show the steps involved in the process. (Fall 2009)
For many applications in behavioral biology, it's nice to trigger events or devices (e.g. cameras, lights) when an animal moves into a specific region of a camera view. In order to trigger our lab's high-speed video cameras, I wrote a simple program in the Java language that uses an inexpensive web-cam to detect if objects move from one frame to the next. A region of interest can be set by clicking and dragging to position a red rectangle on the camera view. Thresholds can be changed to make the trigger more or less sensitive, and it can be toggled between an "armed" and "unarmed" mode.
When exploring sensors for reading out behavioral (i.e. motor) output from an animal, I came across bend sensors that can be incorporated into clothing (http://www.talk2myshirt.com/blog/archives/724). By using thin copper foil, velostat, and epoxy, I was able to shrink the dimensions of the sensor to about 15 mm length. Unfortunatly, I needed something even smaller, and something more compliant. But here's the basic idea:
Beyond sine waves: Perlin noise, images, and, yes, (pixel-based) sinusoids
My second attempt at creating a moving sine wave is somewhat different than in the first example. This time, I created a sine wave class, which during initialization creates an array holding the pixel values of a sine wave. I also added a Perlin noise class, for making a 2D noise pattern, as well as an image class to load an images.
The sequence below was shot in the butterfly house of the Pacific Science Center in Seattle in 2010.
Filmed at 500 fps. This video consists of about 730 frames, which means the whole sequence took less than 1-½ seconds in real time. Shot with a Phantom Miro 4 color high-speed camera, which was graciously loaned to the Daniel Lab by its manufacturer, Vision Research.