Yes fact, with glare described as a Gaussian, extremely bright stars did not look like the ones in your image. With a Gaussian, the edges of the star get sharper as the brightness increases. This matches what we see for the core PSF, but the glare behaves differently.

Here's a very relevant paper titled "Physically-Based Glare Effects for Digital Images": http://citeseerx.ist.psu.edu/viewdoc/do ... 1&type=pdf

The authors describe the PSF for the eye as the sum of three terms: a central Gaussian, a theta^-2 term, and a theta^-3 term (where theta is the angle from the center of the light source.) Here's a sample image of the Southern Cross where Gaussian glare is replaced by G(1/(1 + k*r^3)), where G is the glare brightness, k is a constant that sets the glare falloff rate, and r is the distance in pixels. I deliberately did not the angular distance, as it's more desirable to resolve stars at high zoom factors than it is to try and realistically reproduce all the quirks of a real imaging system (and I realize that's what I did with the diffraction spikes, but they're aesthetically pleasing ).

Global Marketplace For Buying, Selling And Trading Goods And Services

http://www.PublicBuySell.com

## Drawing a better star

krishnafbgroup wrote:Here's a very relevant paper titled "Physically-Based Glare Effects for Digital Images": http://citeseerx.ist.psu.edu/viewdoc/do ... 1&type=pdf

The link is broken, but here's another copy:

http://luthuli.cs.uiuc.edu/~daf/courses/rendering/papers3/spencer95.pdf

### Who is online

Users browsing this forum: 2 guests