On the question of "does the photo look like the real thing" I don’t think there is an easy answer. With long-exposure astrophotos such as the Eskimo Nebula from Kitt Peak, this beautiful shot of NGC 1365 on APOD (credit SSRO), or even my shot on NGC 891, there is always an element of judgement in how the astrophotographer made it look. While you can set standards on star color so your color balance reflects the true spectral colors, judgement in processing I think plays a key role.

We can leave aside narrowband or non-visible electromagnetic raditaion images. Reality vs. perception for those is like guessing what things look like to a bee or a cat.

At the Advanced Imaging Conference in San Jose last November, I listened to a discussion between David Malin (of the Anglo-Australian Observatory) and Jerry Bonnell (of NASA/USRA and co-editor of APOD). The conversation had started from the question "Does the photo look like the real thing?" in reference to astrophotos.

Malin said that, if processed correctly, the image would be the same as if we were able to turn up the magnification and sensitivity of our eyes. If we were out in space in front of the Crab Nebula (M1), we would see what the image shows us.

Bonnell, on the other hand, said that we don’t really know what things would look like if we were there. The density of the light is such that were we close to an nebula, it could be so faint as to be invisible. That even "turning up our eyes" would not necessarily yield the same colors regardless of tuning based on spectra.

My apologies to Malin and Bonnell if I have mistated their positions — this is as I recollect it.

I tend to favor Bonnell’s opinion, although I am certainly not one to pick an argument with Malin! I know that when I put together an astronomical image, I do just that — put it together. Separate images for R/G/B and Luminance, darks and flats applied, sigma reject used to reduce noise and multiple images summed, finally combining into a full-color image. Then comes sharpening with high-pass filters, digitial development, and wavelets, all that affect the relative contrast of objects in the image.

While no data is added, with all these steps I cannot assert a connection to "what it really looks like." I try to make it look pleasing to the eye, but I don’t know if it is accurate. And unless we actually go out there, with probe or spaceship, I don’t think we will know what it looks like.

Certainly good science can be done and we can know many facts about astronomical objects. Visual perception, on the other hand, is so subjective I don’t think we can say what it looks like.