Contrast module and histogram clipping

Questions and answers about processing in StarTools and how to accomplish certain tasks.
Post Reply
gustavo_sanchez
Posts: 23
Joined: Fri Oct 12, 2012 3:37 pm
Location: Puerto Rico
Contact:

Contrast module and histogram clipping

Post by gustavo_sanchez »

Hello,

Usually, after developing a linear image, I went to the Contrast module to try to lower the sky brightness a bit. However, I noticed that after this step the histogram gets moved too much to the left and becomes clipped (as seen later in the Color module). I am supposed to re-Develop my image after the Contrast module, or there is something I am doing wrong? I assumed that StarTools prevented that kind of behavior.
User avatar
admin
Site Admin
Posts: 3364
Joined: Thu Dec 02, 2010 10:51 pm
Location: Melbourne
Contact:

Re: Contrast module and histogram clipping

Post by admin »

gustavo_sanchez wrote:Hello,

Usually, after developing a linear image, I went to the Contrast module to try to lower the sky brightness a bit. However, I noticed that after this step the histogram gets moved too much to the left and becomes clipped (as seen later in the Color module). I am supposed to re-Develop my image after the Contrast module, or there is something I am doing wrong? I assumed that StarTools prevented that kind of behavior.
Hi,

The Contrast module does not clip your image. Where any Dark Anomaly pixel values would become negative (and hence clipped to 0), headroom is re-allocated to such Dark Anomaly pixel values according to the 'Dark Anomaly Headroom' percentage.

For a 16-bit example (0-65535 brightness values), with the 'Dark Anomaly Headroom' percentage at 5%, if the Contrast module encounters a pixel that has a brightness of 128 but has determined that the real background level is 300 (e.g. the pixel it just encountered is darker than the real background and therefore classified as a 'dark anomaly'), then it will 'squeeze' down the anomalous pixel to 5% of its original 128 (= 6) and make the global image's background start at 300 + 6 = 306. This way the anomalous pixels don't disappear or get clipped, they just get less headroom allocated in the image if they come in below the 'real' background level.

If a lot of pixels get classified as a dark anomalies, it may appear in an 8-bit histogram as if the image is getting clipped, but really they are simply squeezed into a smaller amount of headroom. If this happens, and you don't like the result just increase the amount of headroom that they get (re)allocated; you can adjust it to 100% which effectively means they get to retain all headroom they were using up and the global background level is lifted by exactly the amount of the darkest anomaly in your image.

Or you could simply Switch the dark Anomaly filter off if you are certain there are no dark anomalies in your image at all!

Hope this helps,
Ivo Jager
StarTools creator and astronomy enthusiast
Post Reply