New images in the Zoo
As you may already have heard, Galaxy Zoo has new images in it this week!
You may remember my post in September which described how we’ve added images from SDSS’s ‘Stripe 82’. This is an area of the Sloan survey that has been repeatedly imaged to do things like supernova detection (much like that in Supernova Zoo – you have to look at the same place more than once to see what has changed). A benefit of this is that we can add all these images up to make an image that’s like having a much longer exposure than the ordinary SDSS uses.
When the initial batch of these images went in, they did rather stand out a bit, for a few reasons. One was that the colour balance was slightly different to standard SDSS images – in general things looked a little redder than they do usually. Also, another big difference was that sometimes these images had quite a lot of background sky noise. We decided initially to go for this, in order to better make use of the extra faint details we can see – scaling up faint details also scales up the background noise, unfortunately.
We’ve had a second go at this however, and Steven’s developed a clever method of reducing the background noise while not affecting the objects in the image, and we’ve also tweaked the colour balance to make things bluer and fit in more closely with the original SDSS imagery. The background noise gets its colour intensity reduced too, which helps with the sometimes quite intense coloured speckling of the old set. It took us a while to reprocess things, but hopefully this will give us an even better set of data to work with!
You’ll notice quite a few of these images, both from our original Stripe 82 attempt and this new (and we think improved!) set, and we hope you enjoy classifying them and find some gorgeous galaxies out there. There may also be a few artifacts, where the image generation didn’t work out to plan or with more satellites (more images gives more chance for a pesky satellite to get in the way!), so don’t be afraid to hit that artifact button if it looks like there’s a problem.
Have fun classifying!