It only took a few days of collecting data for me to get hooked on this project and also get me searching for a better camera. Not only did the iSight camera images not look all that great, the camera had an annoying tendency to just stop working once or twice a day, for no reason.
After a bit of research, I ordered and received a WiFi-based web camera, the D-Link DCS-930L. Not too expensive ($56), Windows-oriented but from the comments it appeared it could be set up on Mac OSX without much trouble. Same resolution as the iSight, VGA (640×480), but it’s produced much more consistent results and hasn’t failed to send an image to the computer yet. I’ve saved the accumulated images from the iSight but I don’t expect to use them again, as the DCS-930L is so much better. Here’s an example I’ve uploaded to YouTube; these are the images from Sunday, September 9th, taken once per minute, at 10 frames per second.
For a while I was taking the once-per-minute images all night, but I quickly decided that there was not much to see, so I’ve now scheduled the camera to start sending images to the computer about 40 minutes before sunrise until about 40 minutes after sunset.
I’m going to accumulate images for a while until I enough of them to analyze. This is a north facing view, oriented so that it’s never going to see the Sun or the Moon. I’m contemplating getting an additional camera that I could set up to look in different directions, especially west where I can pick up the sunsets.
I’ve started a software project that helps me accumulate and analyze the images as I acquire them. That project is located at Bitbucket.
Eric Floehr’s talk was the initial inspiration for my starting this project. More on his project, which has been ongoing for more than two years, can be found here (links to a PDF).